id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.01883 | Davide D'Ascenzo | Davide D'Ascenzo and Sebastiano Cultrera di Montesano | scDataset: Scalable Data Loading for Deep Learning on Large-Scale Single-Cell Omics | null | null | null | null | cs.LG cs.AI cs.DB q-bio.GN q-bio.QM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Modern single-cell datasets now comprise hundreds of millions of cells, presenting significant challenges for training deep learning models that require shuffled, memory-efficient data loading. While the AnnData format is the community standard for storing single-cell datasets, existing data loading solutions for AnnData are often inadequate: some require loading all data into memory, others convert to dense formats that increase storage demands, and many are hampered by slow random disk access. We present scDataset, a PyTorch IterableDataset that operates directly on one or more AnnData files without the need for format conversion. The core innovation is a combination of block sampling and batched fetching, which together balance randomness and I/O efficiency. On the Tahoe 100M dataset, scDataset achieves up to a 48$\times$ speed-up over AnnLoader, a 27$\times$ speed-up over HuggingFace Datasets, and an 18$\times$ speed-up over BioNeMo in single-core settings. These advances democratize large-scale single-cell model training for the broader research community. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 17:11:49 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"D'Ascenzo",
"Davide",
""
],
[
"di Montesano",
"Sebastiano Cultrera",
""
]
] |
2506.01921 | Minghao Liu Mark | Minghao Liu, Zhitao He, Zhiyuan Fan, Qingyun Wang, Yi R. Fung | MedEBench: Revisiting Text-instructed Image Editing on Medical Domain | Project website: https://mliuby.github.io/MedEBench_Website/ | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Text-guided image editing has seen rapid progress in natural image domains, but its adaptation to medical imaging remains limited and lacks standardized evaluation. Clinically, such editing holds promise for simulating surgical outcomes, creating personalized teaching materials, and enhancing patient communication. To bridge this gap, we introduce MedEBench, a comprehensive benchmark for evaluating text-guided medical image editing. It consists of 1,182 clinically sourced image-prompt triplets spanning 70 tasks across 13 anatomical regions. MedEBench offers three key contributions: (1) a clinically relevant evaluation framework covering Editing Accuracy, Contextual Preservation, and Visual Quality, supported by detailed descriptions of expected change and ROI (Region of Interest) masks; (2) a systematic comparison of seven state-of-the-art models, revealing common failure patterns; and (3) a failure analysis protocol based on attention grounding, using IoU between attention maps and ROIs to identify mislocalization. MedEBench provides a solid foundation for developing and evaluating reliable, clinically meaningful medical image editing systems. Project website: https://mliuby.github.io/MedEBench_Website/ | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 17:43:01 GMT"
},
{
"version": "v2",
"created": "Tue, 03 Jun 2025 12:45:27 GMT"
},
{
"version": "v3",
"created": "Wed, 04 Jun 2025 10:55:29 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Liu",
"Minghao",
""
],
[
"He",
"Zhitao",
""
],
[
"Fan",
"Zhiyuan",
""
],
[
"Wang",
"Qingyun",
""
],
[
"Fung",
"Yi R.",
""
]
] |
2506.01950 | Jiajun Jiang | Jiajun Jiang, Yiming Zhu, Zirui Wu, Jie Song | DualMap: Online Open-Vocabulary Semantic Mapping for Natural Language Navigation in Dynamic Changing Scenes | 8 pages, 5 figures. Code: https://github.com/Eku127/DualMap Project page: https://eku127.github.io/DualMap/ | null | null | null | cs.RO cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce DualMap, an online open-vocabulary mapping system that enables robots to understand and navigate dynamically changing environments through natural language queries. Designed for efficient semantic mapping and adaptability to changing environments, DualMap meets the essential requirements for real-world robot navigation applications. Our proposed hybrid segmentation frontend and object-level status check eliminate the costly 3D object merging required by prior methods, enabling efficient online scene mapping. The dual-map representation combines a global abstract map for high-level candidate selection with a local concrete map for precise goal-reaching, effectively managing and updating dynamic changes in the environment. Through extensive experiments in both simulation and real-world scenarios, we demonstrate state-of-the-art performance in 3D open-vocabulary segmentation, efficient scene mapping, and online language-guided navigation. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 17:59:10 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 17:05:36 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Jiang",
"Jiajun",
""
],
[
"Zhu",
"Yiming",
""
],
[
"Wu",
"Zirui",
""
],
[
"Song",
"Jie",
""
]
] |
2506.01969 | Chang Kong | Pengcuo Dege, Qiuming Luo, Rui Mao, Chang Kong | FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs | 15 pages, conference | null | null | null | cs.DC cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Efficient inference of Multi-Head Latent Attention (MLA) is challenged by deploying the DeepSeek-R1 671B model on a single Multi-GPU server. This paper introduces FlashMLA-ETAP, a novel framework that enhances MLA inference for the single-instance deployment scenario on NVIDIA H20 GPUs. We propose the Efficient Transpose Attention Pipeline (ETAP), which reconfigures attention computation through transposition to align the KV context length with the \(M\)-dimension in WGMMA operations, significantly reducing redundant computations. FlashMLA-ETAP achieves a 2.78x speedup over FlashMLA at 64K sequence length (batch size 16), with 5.24x and 4.94x improvements over FlashAttention-3 and FlashInfer, respectively, while maintaining numerical stability with a 15.2x lower RMSE (\(1.25 \times 10^{-5}\)) than FlashAttention-3. Furthermore, ETAP's design enables seamless integration into frameworks like FlashAttention-3 and FlashInfer, supported by a detailed theoretical analysis. Our work addresses a critical gap in resource-constrained inference, offering a scalable solution for mid-tier GPUs and paving the way for broader adoption in hardware-aware optimization. Code is available at https://github.com/pengcuo/FlashMLA-ETAP. | [
{
"version": "v1",
"created": "Tue, 13 May 2025 17:45:34 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 03:20:26 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Dege",
"Pengcuo",
""
],
[
"Luo",
"Qiuming",
""
],
[
"Mao",
"Rui",
""
],
[
"Kong",
"Chang",
""
]
] |
2506.01982 | Vassilis Lyberatos | Vassilis Lyberatos, Spyridon Kantarelis, Ioanna Zioga, Christina Anagnostopoulou, Giorgos Stamou, Anastasia Georgaki | Music Interpretation and Emotion Perception: A Computational and Neurophysiological Investigation | Accepted at SMC 2025 | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | This study investigates emotional expression and perception in music performance using computational and neurophysiological methods. The influence of different performance settings, such as repertoire, diatonic modal etudes, and improvisation, as well as levels of expressiveness, on performers' emotional communication and listeners' reactions is explored. Professional musicians performed various tasks, and emotional annotations were provided by both performers and the audience. Audio analysis revealed that expressive and improvisational performances exhibited unique acoustic features, while emotion analysis showed stronger emotional responses. Neurophysiological measurements indicated greater relaxation in improvisational performances. This multimodal study highlights the significance of expressivity in enhancing emotional communication and audience engagement. | [
{
"version": "v1",
"created": "Fri, 16 May 2025 15:30:38 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 06:16:47 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Lyberatos",
"Vassilis",
""
],
[
"Kantarelis",
"Spyridon",
""
],
[
"Zioga",
"Ioanna",
""
],
[
"Anagnostopoulou",
"Christina",
""
],
[
"Stamou",
"Giorgos",
""
],
[
"Georgaki",
"Anastasia",
""
]
] |
2506.01991 | Mohammad Fakhruddin Babar | Mohammad Fakhruddin Babar, Zain A. H. Hammadeh, Mohammad Hamad, and Monowar Hasan | Investigating Timing-Based Information Leakage in Data Flow-Driven Real-Time Systems | null | null | null | null | cs.DC cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | Leaking information about the execution behavior of critical real-time tasks may lead to serious consequences, including violations of temporal constraints and even severe failures. We study information leakage for a special class of real-time tasks that have two execution modes, namely, typical execution (which invokes the majority of times) and critical execution (to tackle exceptional conditions). The data flow-driven applications inherit such a multimode execution model. In this paper, we investigate whether a low-priority "observer" task can infer the execution patterns of a high-priority "victim" task (especially the critical executions). We develop a new statistical analysis technique and show that by analyzing the response times of the low-priority task, it becomes possible to extract the execution behavior of the high-priority task. We test our approach against a random selection technique that arbitrarily classifies a job as critical. We find that correlating the observer's response times with the victim's jobs can result in higher precision in identifying critical invocations compared to a random guess. We conduct extensive evaluations with systemically generated workloads, including a case study using a UAV autopilot (ArduPilot) taskset parameters. We found that our inference algorithm can achieve relatively low false positive rates (less than 25%) with relatively low footprint (1 MB memory and 50 ms timing overhead on a Raspberry Pi 4 platform). We further demonstrate the feasibility of inference on two cyber-physical platforms: an off-the-shelf manufacturing robot and a custom-built surveillance system. | [
{
"version": "v1",
"created": "Sun, 18 May 2025 03:39:44 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:39:58 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Babar",
"Mohammad Fakhruddin",
""
],
[
"Hammadeh",
"Zain A. H.",
""
],
[
"Hamad",
"Mohammad",
""
],
[
"Hasan",
"Monowar",
""
]
] |
2506.02005 | Timothy Do | Timothy Do, Pranav Saran, Harshita Poojary, Pranav Prabhu, Sean O'Brien, Vasu Sharma, Kevin Zhu | Pruning for Performance: Efficient Idiom and Metaphor Classification in Low-Resource Konkani Using mBERT | 9 pages, 7 figures | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | In this paper, we address the persistent challenges that figurative language expressions pose for natural language processing (NLP) systems, particularly in low-resource languages such as Konkani. We present a hybrid model that integrates a pre-trained Multilingual BERT (mBERT) with a bidirectional LSTM and a linear classifier. This architecture is fine-tuned on a newly introduced annotated dataset for metaphor classification, developed as part of this work. To improve the model's efficiency, we implement a gradient-based attention head pruning strategy. For metaphor classification, the pruned model achieves an accuracy of 78%. We also applied our pruning approach to expand on an existing idiom classification task, achieving 83% accuracy. These results demonstrate the effectiveness of attention head pruning for building efficient NLP tools in underrepresented languages. | [
{
"version": "v1",
"created": "Sat, 24 May 2025 03:23:39 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Do",
"Timothy",
""
],
[
"Saran",
"Pranav",
""
],
[
"Poojary",
"Harshita",
""
],
[
"Prabhu",
"Pranav",
""
],
[
"O'Brien",
"Sean",
""
],
[
"Sharma",
"Vasu",
""
],
[
"Zhu",
"Kevin",
""
]
] |
2506.02031 | Matthew Rayman | Matthew Rayman | Effective Versions of Strong Measure Zero | null | null | null | null | math.LO cs.CC cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Effective versions of strong measure zero sets are developed for various levels of complexity and computability. It is shown that the sets can be equivalently defined using a generalization of supermartingales called odds supermartingales, success rates on supermartingales, predictors, and coverings. We show Borel's conjecture of a set having strong measure zero if and only if it is countable holds in the time and space bounded setting. At the level of computability this does not hold. We show the computable level contains sequences at arbitrary levels of the hyperarithmetical hierarchy by proving a correspondence principle yielding a condition for the sets of computable strong measure zero to agree with the classical sets of strong measure zero. An algorithmic version of strong measure zero using lower semicomputability is defined. We show that this notion is equivalent to the set of NCR reals studied by Reimann and Slaman, thereby giving new characterizations of this set. Effective strong packing dimension zero is investigated requiring success with respect to the limit inferior instead of the limit superior. It is proven that every sequence in the corresponding algorithmic class is decidable. At the level of computability, the sets coincide with a notion of weak countability that we define. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 17:23:47 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Rayman",
"Matthew",
""
]
] |
2506.02112 | Tian Xia | Xuweiyi Chen, Tian Xia, Sihan Xu, Jianing Yang, Joyce Chai, Zezhou Cheng | SAB3R: Semantic-Augmented Backbone in 3D Reconstruction | 3D-LLM/VLA @ CVPR2025 | Project page: https://uva-computer-vision-lab.github.io/sab3r/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We introduce a new task, Map and Locate, which unifies the traditionally distinct objectives of open-vocabulary segmentation - detecting and segmenting object instances based on natural language queries - and 3D reconstruction, the process of estimating a scene's 3D structure from visual inputs. Specifically, Map and Locate involves generating a point cloud from an unposed video and segmenting object instances based on open-vocabulary queries. This task serves as a critical step toward real-world embodied AI applications and introduces a practical task that bridges reconstruction, recognition and reorganization. To tackle this task, we introduce a simple yet effective baseline, which we denote as SAB3R. Our approach builds upon MASt3R, a recent breakthrough in 3D computer vision, and incorporates a lightweight distillation strategy. This method transfers dense, per-pixel semantic features from 2D vision backbones (eg, CLIP and DINOv2) to enhance MASt3R's capabilities. Without introducing any auxiliary frozen networks, our model generates per-pixel semantic features and constructs cohesive point maps in a single forward pass. Compared to separately deploying MASt3R and CLIP, our unified model, SAB3R, achieves superior performance on the Map and Locate benchmark. Furthermore, we evaluate SAB3R on both 2D semantic segmentation and 3D tasks to comprehensively validate its effectiveness. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 18:00:04 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:28:08 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chen",
"Xuweiyi",
""
],
[
"Xia",
"Tian",
""
],
[
"Xu",
"Sihan",
""
],
[
"Yang",
"Jianing",
""
],
[
"Chai",
"Joyce",
""
],
[
"Cheng",
"Zezhou",
""
]
] |
2506.02120 | Mauricio Resende | Mariana A. Londe and Luciana S. Pessoa and Carlos E. Andrade and Jos\'e F. Gon\c{c}alves and Mauricio G. C. Resende | Random-key genetic algorithms: Principles and applications | 21 pages, 1 figure, 1 table, 1 algorithm, forthcoming in Handbook of Heuristics, 2nd edition, SpringerNature, New York | null | null | null | cs.NE cs.AI math.OC | http://creativecommons.org/licenses/by/4.0/ | A random-key genetic algorithm is an evolutionary metaheuristic for discrete and global optimization. Each solution is encoded as a vector of N random keys, where a random key is a real number randomly generated in the continuous interval [0, 1). A decoder maps each vector of random keys to a solution of the optimization problem being solved and computes its cost. The benefit of this approach is that all genetic operators and transformations can be maintained within the unitary hypercube, regardless of the problem being addressed. This enhances the productivity and maintainability of the core framework. The algorithm starts with a population of P vectors of random keys. At each iteration, the vectors are partitioned into two sets: a smaller set of high-valued elite solutions and the remaining non-elite solutions. All elite elements are copied, without change, to the next population. A small number of random-key vectors (the mutants) is added to the population of the next iteration. The remaining elements of the population of the next iteration are generated by combining, with the parametrized uniform crossover of Spears and DeJong (1991), pairs of solutions. This chapter reviews random-key genetic algorithms and describes an effective variant called biased random-key genetic algorithms. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 18:00:07 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 17:44:05 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Londe",
"Mariana A.",
""
],
[
"Pessoa",
"Luciana S.",
""
],
[
"Andrade",
"Carlos E.",
""
],
[
"Gonçalves",
"José F.",
""
],
[
"Resende",
"Mauricio G. C.",
""
]
] |
2506.02132 | Michael Li | Michael Li, Nishant Subramani | Model Internal Sleuthing: Finding Lexical Identity and Inflectional Morphology in Modern Language Models | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large transformer-based language models dominate modern NLP, yet our understanding of how they encode linguistic information is rooted in studies of early models like BERT and GPT-2. To better understand today's language models, we investigate how both classical architectures (BERT, DeBERTa, GPT-2)and contemporary large language models (Pythia, OLMo-2, Gemma-2, Qwen2.5, Llama-3.1) represent lexical identity and inflectional morphology. We train linear and nonlinear classifiers on layer-wise activations to predict word lemmas and inflectional features. We discover that models concentrate lexical information linearly in early layers and increasingly nonlinearly in later layers, while keeping inflectional information uniformly accessible and linearly separable throughout the layers. Further analysis reveals that these models encode inflectional morphology through generalizable abstractions, but rely predominantly on memorization to encode lexical identity. Remarkably, these patterns emerge across all 16 models we test, despite differences in architecture, size, and training regime (including pretrained and instruction-tuned variants). This consistency suggests that, despite substantial advances in LLM technologies, transformer models organize linguistic information in similar ways, indicating that these properties could be fundamental for next token prediction and are learned early during pretraining. Our code is available at https://github.com/ml5885/model_internal_sleuthing | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 18:01:56 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:03:49 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Li",
"Michael",
""
],
[
"Subramani",
"Nishant",
""
]
] |
2506.02139 | Edward Chang | Edward Y. Chang | The Unified Cognitive Consciousness Theory for Language Models: Anchoring Semantics, Thresholds of Activation, and Emergent Reasoning | 12 pages, 2 figure, 1 table | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot learning in large language models (LLMs) reveals a core paradox: certain tasks generalize from just a few examples, while others demand extensive supervision. To explain this, we introduce the Unified Cognitive Consciousness Theory (UCCT), which reconceptualizes LLMs not as deficient agents, but as unconscious substrates: dense, distributed repositories of linguistic and conceptual patterns that operate without explicit semantics, intention, or goal-directed reasoning. Under this view, LLMs are not flawed simulations of cognition but foundational substrates for general intelligence. UCCT posits that semantic anchoring, via prompts, role assignments, and structured interaction, functions as a conscious control layer that modulates latent representations toward task-relevant semantics and enables coherent, structured reasoning. It unifies prompting, fine-tuning, retrieval-augmented generalization, and multi-agent collaboration within a single framework, grounded in the probabilistic alignment between unconscious pattern space and externally imposed semantic constraints (e.g., prompts, supervision, task objectives). The core implication is not to replace LLMs, but to integrate and unify them through a structured cognitive layer that supports intentional reasoning. This enables collections of LLMs to operate within domain-specialized verticals (e.g., legal reasoning, medical diagnosis) that reason, regulate, and adapt together. Such integration is characterized by phase-transition behavior, wherein anchored representations cross coherence thresholds as a function of semantic constraint strength and interaction context. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 18:12:43 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:44:46 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chang",
"Edward Y.",
""
]
] |
2506.02197 | Marcos V. Conde | Marcos V. Conde and Radu Timofte and Zihao Lu and Xiangyu Kong and Xiaoxia Xing and Fan Wang and Suejin Han and MinKyu Park and Tianyu Zhang and Xin Luo and Yeda Chen and Dong Liu and Li Pang and Yuhang Yang and Hongzhong Wang and Xiangyong Cao and Ruixuan Jiang and Senyan Xu and Siyuan Jiang and Xueyang Fu and Zheng-Jun Zha and Tianyu Hao and Yuhong He and Ruoqi Li and Yueqi Yang and Xiang Yu and Guanlan Hong and Minmin Yi and Yuanjia Chen and Liwen Zhang and Zijie Jin and Cheng Li and Lian Liu and Wei Song and Heng Sun and Yubo Wang and Jinghua Wang and Jiajie Lu and Watchara Ruangsan | NTIRE 2025 Challenge on RAW Image Restoration and Super-Resolution | CVPR 2025 - New Trends in Image Restoration and Enhancement (NTIRE) | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This paper reviews the NTIRE 2025 RAW Image Restoration and Super-Resolution Challenge, highlighting the proposed solutions and results. New methods for RAW Restoration and Super-Resolution could be essential in modern Image Signal Processing (ISP) pipelines, however, this problem is not as explored as in the RGB domain. The goal of this challenge is two fold, (i) restore RAW images with blur and noise degradations, (ii) upscale RAW Bayer images by 2x, considering unknown noise and blur. In the challenge, a total of 230 participants registered, and 45 submitted results during thee challenge period. This report presents the current state-of-the-art in RAW Restoration. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 19:34:21 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 13:04:27 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Conde",
"Marcos V.",
""
],
[
"Timofte",
"Radu",
""
],
[
"Lu",
"Zihao",
""
],
[
"Kong",
"Xiangyu",
""
],
[
"Xing",
"Xiaoxia",
""
],
[
"Wang",
"Fan",
""
],
[
"Han",
"Suejin",
""
],
[
"Park",
"MinKyu",
""
],
[
"Zhang",
"Tianyu",
""
],
[
"Luo",
"Xin",
""
],
[
"Chen",
"Yeda",
""
],
[
"Liu",
"Dong",
""
],
[
"Pang",
"Li",
""
],
[
"Yang",
"Yuhang",
""
],
[
"Wang",
"Hongzhong",
""
],
[
"Cao",
"Xiangyong",
""
],
[
"Jiang",
"Ruixuan",
""
],
[
"Xu",
"Senyan",
""
],
[
"Jiang",
"Siyuan",
""
],
[
"Fu",
"Xueyang",
""
],
[
"Zha",
"Zheng-Jun",
""
],
[
"Hao",
"Tianyu",
""
],
[
"He",
"Yuhong",
""
],
[
"Li",
"Ruoqi",
""
],
[
"Yang",
"Yueqi",
""
],
[
"Yu",
"Xiang",
""
],
[
"Hong",
"Guanlan",
""
],
[
"Yi",
"Minmin",
""
],
[
"Chen",
"Yuanjia",
""
],
[
"Zhang",
"Liwen",
""
],
[
"Jin",
"Zijie",
""
],
[
"Li",
"Cheng",
""
],
[
"Liu",
"Lian",
""
],
[
"Song",
"Wei",
""
],
[
"Sun",
"Heng",
""
],
[
"Wang",
"Yubo",
""
],
[
"Wang",
"Jinghua",
""
],
[
"Lu",
"Jiajie",
""
],
[
"Ruangsan",
"Watchara",
""
]
] |
2506.02290 | Zhan Song | Jiaqi Yin, Zhan Song, Nicolas Bohm Agostini, Antonino Tumeo, Cunxi Yu | HEC: Equivalence Verification Checking for Code Transformation via Equality Saturation | Accepted by USENIX ATC 2025 | null | null | null | cs.AR cs.PL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In modern computing systems, compilation employs numerous optimization techniques to enhance code performance. Source-to-source code transformations, which include control flow and datapath transformations, have been widely used in High-Level Synthesis (HLS) and compiler optimization.
While researchers actively investigate methods to improve performance with source-to-source code transformations, they often overlook the significance of verifying their correctness. Current tools cannot provide a holistic verification of these transformations. This paper introduces HEC, a framework for equivalence checking that leverages the e-graph data structure to comprehensively verify functional equivalence between programs. HEC utilizes the MLIR as its frontend and integrates MLIR into the e-graph framework. Through the combination of dynamic and static e-graph rewriting, HEC facilitates the validation of comprehensive code transformations.
We demonstrate effectiveness of HEC on PolyBenchC benchmarks, successfully verifying loop unrolling, tiling, and fusion transformations. HEC processes over 100,000 lines of MLIR code in 40 minutes with predictable runtime scaling. Importantly, HEC identified two critical compilation errors in mlir-opt: loop boundary check errors causing unintended executions during unrolling, and memory read-after-write violations in loop fusion that alter program semantics. These findings demonstrate HEC practical value in detecting real-world compiler bugs and highlight the importance of formal verification in optimization pipelines. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 21:59:17 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yin",
"Jiaqi",
""
],
[
"Song",
"Zhan",
""
],
[
"Agostini",
"Nicolas Bohm",
""
],
[
"Tumeo",
"Antonino",
""
],
[
"Yu",
"Cunxi",
""
]
] |
2506.02294 | Niclas Popp | Niclas Popp, Kevin Alexander Laube, Matthias Hein, Lukas Schott | Improving Knowledge Distillation Under Unknown Covariate Shift Through Confidence-Guided Data Augmentation | null | null | null | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | Large foundation models trained on extensive datasets demonstrate strong zero-shot capabilities in various domains. To replicate their success when data and model size are constrained, knowledge distillation has become an established tool for transferring knowledge from foundation models to small student networks. However, the effectiveness of distillation is critically limited by the available training data. This work addresses the common practical issue of covariate shift in knowledge distillation, where spurious features appear during training but not at test time. We ask the question: when these spurious features are unknown, yet a robust teacher is available, is it possible for a student to also become robust to them? We address this problem by introducing a novel diffusion-based data augmentation strategy that generates images by maximizing the disagreement between the teacher and the student, effectively creating challenging samples that the student struggles with. Experiments demonstrate that our approach significantly improves worst group and mean group accuracy on CelebA and SpuCo Birds as well as the spurious mAUC on spurious ImageNet under covariate shift, outperforming state-of-the-art diffusion-based data augmentation baselines | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 22:15:59 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 01:55:38 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Popp",
"Niclas",
""
],
[
"Laube",
"Kevin Alexander",
""
],
[
"Hein",
"Matthias",
""
],
[
"Schott",
"Lukas",
""
]
] |
2506.02308 | Xiaojun Shan | Xiaojun Shan, Qi Cao, Xing Han, Haofei Yu, Paul Pu Liang | MINT: Multimodal Instruction Tuning with Multimodal Interaction Grouping | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent advances in multimodal foundation models have achieved state-of-the-art performance across a range of tasks. These breakthroughs are largely driven by new pre-training paradigms that leverage large-scale, unlabeled multimodal data, followed by instruction fine-tuning on curated labeled datasets and high-quality prompts. While there is growing interest in scaling instruction fine-tuning to ever-larger datasets in both quantity and scale, our findings reveal that simply increasing the number of instruction-tuning tasks does not consistently yield better performance. Instead, we observe that grouping tasks by the common interactions across modalities, such as discovering redundant shared information, prioritizing modality selection with unique information, or requiring synergistic fusion to discover new information from both modalities, encourages the models to learn transferrable skills within a group while suppressing interference from mismatched tasks. To this end, we introduce MINT, a simple yet surprisingly effective task-grouping strategy based on the type of multimodal interaction. We demonstrate that the proposed method greatly outperforms existing task grouping baselines for multimodal instruction tuning, striking an effective balance between generalization and specialization. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 22:55:23 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:54:13 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Shan",
"Xiaojun",
""
],
[
"Cao",
"Qi",
""
],
[
"Han",
"Xing",
""
],
[
"Yu",
"Haofei",
""
],
[
"Liang",
"Paul Pu",
""
]
] |
2506.02356 | Woojeong Jin | Woojeong Jin, Seongchan Kim, Seungryong Kim | InterRVOS: Interaction-aware Referring Video Object Segmentation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Referring video object segmentation aims to segment the object in a video corresponding to a given natural language expression. While prior works have explored various referring scenarios, including motion-centric or multi-instance expressions, most approaches still focus on localizing a single target object in isolation. However, in comprehensive video understanding, an object's role is often defined by its interactions with other entities, which are largely overlooked in existing datasets and models. In this work, we introduce Interaction-aware referring video object sgementation (InterRVOS), a new task that requires segmenting both actor and target entities involved in an interaction. Each interactoin is described through a pair of complementary expressions from different semantic perspectives, enabling fine-grained modeling of inter-object relationships. To tackle this task, we propose InterRVOS-8K, the large-scale and automatically constructed dataset containing diverse interaction-aware expressions with corresponding masks, including challenging cases such as motion-only multi-instance expressions. We also present a baseline architecture, ReVIOSa, designed to handle actor-target segmentation from a single expression, achieving strong performance in both standard and interaction-focused settings. Furthermore, we introduce an actor-target-aware evalaution setting that enables a more targeted assessment of interaction understanding. Experimental results demonstrate that our approach outperforms prior methods in modeling complex object interactions for referring video object segmentation task, establishing a strong foundation for future research in interaction-centric video understanding. Our project page is available at https://cvlab-kaist.github.io/InterRVOS. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:16:13 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 09:23:48 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Jin",
"Woojeong",
""
],
[
"Kim",
"Seongchan",
""
],
[
"Kim",
"Seungryong",
""
]
] |
2506.02426 | Maryam Berijanian | Maryam Berijanian and Kuldeep Singh and Amin Sehati | Comparative Analysis of AI Agent Architectures for Entity Relationship Classification | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Entity relationship classification remains a challenging task in information extraction, especially in scenarios with limited labeled data and complex relational structures. In this study, we conduct a comparative analysis of three distinct AI agent architectures designed to perform relation classification using large language models (LLMs). The agentic architectures explored include (1) reflective self-evaluation, (2) hierarchical task decomposition, and (3) a novel multi-agent dynamic example generation mechanism, each leveraging different modes of reasoning and prompt adaptation. In particular, our dynamic example generation approach introduces real-time cooperative and adversarial prompting. We systematically compare their performance across multiple domains and model backends. Our experiments demonstrate that multi-agent coordination consistently outperforms standard few-shot prompting and approaches the performance of fine-tuned models. These findings offer practical guidance for the design of modular, generalizable LLM-based systems for structured relation extraction. The source codes and dataset are available at https://github.com/maryambrj/ALIEN.git. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:19:47 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 14:21:02 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Berijanian",
"Maryam",
""
],
[
"Singh",
"Kuldeep",
""
],
[
"Sehati",
"Amin",
""
]
] |
2506.02427 | Mini Das | Jingcheng Yuan and Mini Das | Single-Shot, Single-Mask X-ray Dark-field and Phase Contrast Imaging | 18 pages, 10 figures | null | null | null | physics.med-ph physics.app-ph physics.ins-det physics.optics | http://creativecommons.org/licenses/by/4.0/ | X-ray imaging, traditionally relying on attenuation contrast, struggles to differentiate materials with similar attenuation coefficients like soft tissues. X-ray phase contrast imaging (XPCI) and dark-field (DF) imaging provide enhanced contrast by detecting phase shifts and ultra-small-angle X-ray scattering (USAXS). However, they typically require complex and costly setups, along with multiple exposures to retrieve various contrast features. In this study, we introduce a novel single-mask X-ray imaging system design that simultaneously captures attenuation, differential phase contrast (DPC), and dark-field images in a single exposure. Most importantly, our proposed system design requires just a single mask alignment with relatively low-resolution detectors. Using our novel light transport models derived for these specific system designs, we show intuitive understanding of contrast formation and retrieval method of different contrast features. Our approach eliminates the need for highly coherent X-ray sources, ultra-high-resolution detectors, spectral detectors or intricate gratings. We propose three variations of the single-mask setup, each optimized for different contrast types, offering flexibility and efficiency in a variety of applications. The versatility of this single-mask approach along with the use of befitting light transport models holds promise for broader use in clinical diagnostics and industrial inspection, making advanced X-ray imaging more accessible and cost-effective. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:21:38 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yuan",
"Jingcheng",
""
],
[
"Das",
"Mini",
""
]
] |
2506.02442 | Utsav Maskey | Utsav Maskey, Mark Dras, Usman Naseem | Should LLM Safety Be More Than Refusing Harmful Instructions? | Preprint | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This paper presents a systematic evaluation of Large Language Models' (LLMs) behavior on long-tail distributed (encrypted) texts and their safety implications. We introduce a two-dimensional framework for assessing LLM safety: (1) instruction refusal-the ability to reject harmful obfuscated instructions, and (2) generation safety-the suppression of generating harmful responses. Through comprehensive experiments, we demonstrate that models that possess capabilities to decrypt ciphers may be susceptible to mismatched-generalization attacks: their safety mechanisms fail on at least one safety dimension, leading to unsafe responses or over-refusal. Based on these findings, we evaluate a number of pre-LLM and post-LLM safeguards and discuss their strengths and limitations. This work contributes to understanding the safety of LLM in long-tail text scenarios and provides directions for developing robust safety mechanisms. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:00:12 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 05:56:40 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Maskey",
"Utsav",
""
],
[
"Dras",
"Mark",
""
],
[
"Naseem",
"Usman",
""
]
] |
2506.02491 | Guangwu Xu | Guangwu Xu, Yunxiao Tian, and Bingxin Yang | On the Inversion Modulo a Power of an Integer | null | null | null | null | cs.DS | http://creativecommons.org/licenses/by/4.0/ | Recently, Koc proposed a neat and efficient algorithm for computing $x = a^{-1} \pmod {p^k}$ for a prime $p$ based on the exact solution of linear equations using $p$-adic expansions. The algorithm requires only addition and right shift per step. In this paper, we design an algorithm that computes $x = a^{-1} \pmod {n^k}$ for any integer $n>1$. The algorithm has a motivation from the schoolbook multiplication and achieves both efficiency and generality. The greater flexibility of our algorithm is explored by utilizing the build-in arithmetic of computer architecture, e.g., $n=2^{64}$, and experimental results show significant improvements. This paper also contains some results on modular inverse based on an alternative proof of correctness of Koc algorithm. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 06:12:54 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 05:16:52 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Xu",
"Guangwu",
""
],
[
"Tian",
"Yunxiao",
""
],
[
"Yang",
"Bingxin",
""
]
] |
2506.02535 | Juntong Li | Juntong Li, Lingwei Dang, Yukun Su, Yun Hao, Qingxin Xiao, Yongwei Nie, Qingyao Wu | MemoryOut: Learning Principal Features via Multimodal Sparse Filtering Network for Semi-supervised Video Anomaly Detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video Anomaly Detection (VAD) methods based on reconstruction or prediction face two critical challenges: (1) strong generalization capability often results in accurate reconstruction or prediction of abnormal events, making it difficult to distinguish normal from abnormal patterns; (2) reliance only on low-level appearance and motion cues limits their ability to identify high-level semantic in abnormal events from complex scenes. To address these limitations, we propose a novel VAD framework with two key innovations. First, to suppress excessive generalization, we introduce the Sparse Feature Filtering Module (SFFM) that employs bottleneck filters to dynamically and adaptively remove abnormal information from features. Unlike traditional memory modules, it does not need to memorize the normal prototypes across the training dataset. Further, we design the Mixture of Experts (MoE) architecture for SFFM. Each expert is responsible for extracting specialized principal features during running time, and different experts are selectively activated to ensure the diversity of the learned principal features. Second, to overcome the neglect of semantics in existing methods, we integrate a Vision-Language Model (VLM) to generate textual descriptions for video clips, enabling comprehensive joint modeling of semantic, appearance, and motion cues. Additionally, we enforce modality consistency through semantic similarity constraints and motion frame-difference contrastive loss. Extensive experiments on multiple public datasets validate the effectiveness of our multimodal joint modeling framework and sparse feature filtering paradigm. Project page at https://qzfm.github.io/sfn_vad_project_page/. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 07:14:57 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 06:36:25 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Li",
"Juntong",
""
],
[
"Dang",
"Lingwei",
""
],
[
"Su",
"Yukun",
""
],
[
"Hao",
"Yun",
""
],
[
"Xiao",
"Qingxin",
""
],
[
"Nie",
"Yongwei",
""
],
[
"Wu",
"Qingyao",
""
]
] |
2506.02551 | Xinghuan Wang | Xinghuan Wang, Xiangcun Meng | Impact of convective overshooting on the single-degenerate model of Type Ia supernovae | 11pages, 9 figures | null | 10.1051/0004-6361/202554449 | null | astro-ph.SR astro-ph.HE | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The single-degenerate (SD) model is one of the principal models for the progenitors of Type Ia supernovae (SNe Ia). However, it faces some challenges, the primary being its inability to account for the observed SN Ia birth rate. Many studies have attempted to address this issue by expanding the parameter space, defined by the initial donor star mass and orbital period, that can lead to SNe Ia, as well as by improving binary population synthesis. While these efforts have led to significant progress, many uncertainties in stellar physics persist, which influences the outcomes of such studies. Convective overshooting, which can significantly affect the internal structure of a star and subsequently its evolution within a binary system, is one of the most significant sources of uncertainty in stellar physics. We investigate the effect of convective overshooting on the parameter space and birth rate of SNe Ia within the SD model. We employed the common-envelope wind (CEW) model, a new version of the SD model, as our progenitor model. Using MESA, we obtained the parameter space that leads to SNe Ia for three different convective overshooting parameters and calculated the corresponding SN Ia birth rate. Convective overshooting expands the upper boundaries (corresponding to a larger initial donor mass) and right boundaries (corresponding to a longer initial orbital period) of the parameter space for systems with massive white dwarfs (WDs; >= 0.75Msun). However, the minimum WD mass and the parameter space for low-mass WDs - and, consequently, the calculated SN Ia birth rate - vary non-monotonically with convective overshooting parameters. The CEW model may explain the SNe Ia that interact with the circumstellar medium (CSM), i.e., SNe Ia-CSM. We find that the parameter space for SNe Ia-CSM increases with convective overshooting parameters, as does their birth rate. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 07:38:54 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 03:26:28 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wang",
"Xinghuan",
""
],
[
"Meng",
"Xiangcun",
""
]
] |
2506.02606 | Baoyang Chen | Baoyang Chen, Xian Xu, Huamin Qu | Multi Layered Autonomy and AI Ecologies in Robotic Art Installations | null | null | 10.48550/arXiv.2410.22462 | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Symbiosis of Agents is a large-scale installation by Baoyang Chen (baoyangchen.com) that embeds AI-driven robots in an immersive, mirror-lined arena, probing the tension between machine agency and artistic authorship. Drawing on early cybernetics, rule-based conceptual art, and seminal robotic works, it orchestrates fluid exchanges among robotic arms, quadruped machines, their environment, and the public. A three tier faith system pilots the ecology: micro-level adaptive tactics, meso-level narrative drives, and a macro-level prime directive. This hierarchy lets behaviors evolve organically in response to environmental cues and even a viewer's breath, turning spectators into co-authors of the unfolding drama. Framed by a speculative terraforming scenario that recalls the historical exploitation of marginalized labor, the piece asks who bears responsibility in AI-mediated futures. Choreographed motion, AI-generated scripts, reactive lighting, and drifting fog cast the robots as collaborators rather than tools, forging a living, emergent artwork. Exhibited internationally, Symbiosis of Agents shows how cybernetic feedback, robotic experimentation, and conceptual rule-making can converge to redefine agency, authorship, and ethics in contemporary art. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 08:28:19 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 04:57:32 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chen",
"Baoyang",
""
],
[
"Xu",
"Xian",
""
],
[
"Qu",
"Huamin",
""
]
] |
2506.02614 | Guohang Zhuang | Guohang Zhuang, Weixi Song, Jinyang Huang, Chenwei Yang, Yan Lu | High Performance Space Debris Tracking in Complex Skylight Backgrounds with a Large-Scale Dataset | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | With the rapid development of space exploration, space debris has attracted more attention due to its potential extreme threat, leading to the need for real-time and accurate debris tracking. However, existing methods are mainly based on traditional signal processing, which cannot effectively process the complex background and dense space debris. In this paper, we propose a deep learning-based Space Debris Tracking Network~(SDT-Net) to achieve highly accurate debris tracking. SDT-Net effectively represents the feature of debris, enhancing the efficiency and stability of end-to-end model learning. To train and evaluate this model effectively, we also produce a large-scale dataset Space Debris Tracking Dataset (SDTD) by a novel observation-based data simulation scheme. SDTD contains 18,040 video sequences with a total of 62,562 frames and covers 250,000 synthetic space debris. Extensive experiments validate the effectiveness of our model and the challenging of our dataset. Furthermore, we test our model on real data from the Antarctic Station, achieving a MOTA score of 70.6%, which demonstrates its strong transferability to real-world scenarios. Our dataset and code will be released soon. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 08:30:25 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 08:10:38 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zhuang",
"Guohang",
""
],
[
"Song",
"Weixi",
""
],
[
"Huang",
"Jinyang",
""
],
[
"Yang",
"Chenwei",
""
],
[
"Lu",
"Yan",
""
]
] |
2506.02635 | Jannis Halbey | Jannis Halbey, Seta Rakotomandimby, Mathieu Besan\c{c}on, S\'ebastien Designolle and Sebastian Pokutta | Efficient Quadratic Corrections for Frank-Wolfe Algorithms | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | We develop a Frank-Wolfe algorithm with corrective steps, generalizing previous algorithms including blended pairwise conditional gradients and fully-corrective Frank-Wolfe, and propose a highly efficient corrective algorithm in the case of convex quadratic objectives based on linear optimization or linear system solving, akin to Wolfe's minimum-norm point. Beyond optimization problems that are directly quadratic, we revisit two algorithms, split conditional gradient for the intersection of two sets accessible through linear oracles, and second-order conditional gradient sliding, which approximately solves Variable-Metric projection subproblems, proposing improvement of the algorithms and their guarantees that may be of interest beyond our work, and we leverage quadratic corrections to accelerate the quadratic subproblems they solve. We show significant computational acceleration of Frank-Wolfe-based algorithms equipped with the quadratic correction on the considered problem classes. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 08:52:24 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 07:52:40 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Halbey",
"Jannis",
""
],
[
"Rakotomandimby",
"Seta",
""
],
[
"Besançon",
"Mathieu",
""
],
[
"Designolle",
"Sébastien",
""
],
[
"Pokutta",
"Sebastian",
""
]
] |
2506.02644 | Hidekazu Yoshioka | H. Yoshioka, M. Tsujimura, T. Tanaka | Non-exchangeable evolutionary and mean field games and their applications | null | null | null | null | math.OC cs.NA math.NA math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A replicator dynamic for non-exchangeable agents in a continuous action space is formulated and its well-posedness is proven in a space of probability measures. The non-exchangeability allows for the analysis of evolutionary games involving agents with distinct (and possibly infinitely many) types. We also explicitly connect this replicator dynamic to a stationary mean field game, which determines the pairwise actions of the heterogeneous agents. Moreover, as a byproduct of our theoretical results, we show that a class of nonlinear voter models, recently the subject of increasing interest, called q-voter models, can be viewed as a replicator dynamic driven by a utility that is a power of the probability density. This implies that non-exchangeable and/or mean-field game formulations of these models can also be constructed. We also present computational examples of evolutionary and mean field game models using a finite difference method, focusing on tragedy of the commons and the q-voter model with non-exchangeable agents, of which are interesting cases from theoretical and computational perspectives. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:00:03 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 05:22:18 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yoshioka",
"H.",
""
],
[
"Tsujimura",
"M.",
""
],
[
"Tanaka",
"T.",
""
]
] |
2506.02651 | Luca Arnaboldi | Luca Arnaboldi, Bruno Loureiro, Ludovic Stephan, Florent Krzakala, Lenka Zdeborova | Asymptotics of SGD in Sequence-Single Index Models and Single-Layer Attention Networks | null | null | null | null | stat.ML cond-mat.dis-nn cs.LG | http://creativecommons.org/licenses/by/4.0/ | We study the dynamics of stochastic gradient descent (SGD) for a class of sequence models termed Sequence Single-Index (SSI) models, where the target depends on a single direction in input space applied to a sequence of tokens. This setting generalizes classical single-index models to the sequential domain, encompassing simplified one-layer attention architectures. We derive a closed-form expression for the population loss in terms of a pair of sufficient statistics capturing semantic and positional alignment, and characterize the induced high-dimensional SGD dynamics for these coordinates. Our analysis reveals two distinct training phases: escape from uninformative initialization and alignment with the target subspace, and demonstrates how the sequence length and positional encoding influence convergence speed and learning trajectories. These results provide a rigorous and interpretable foundation for understanding how sequential structure in data can be beneficial for learning with attention-based models. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:03:27 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Arnaboldi",
"Luca",
""
],
[
"Loureiro",
"Bruno",
""
],
[
"Stephan",
"Ludovic",
""
],
[
"Krzakala",
"Florent",
""
],
[
"Zdeborova",
"Lenka",
""
]
] |
2506.02658 | Kechi Zhang | Kechi Zhang, Ge Li, Jia Li, Huangzhao Zhang, Jingjing Xu, Hao Zhu, Lecheng Wang, Jia Li, Yihong Dong, Jing Mai, Bin Gu, Zhi Jin | Computational Thinking Reasoning in Large Language Models | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While large language models (LLMs) have demonstrated remarkable reasoning capabilities, they often struggle with complex tasks that require specific thinking paradigms, such as divide-and-conquer and procedural deduction, \etc Previous researches integrate external, reliable tools to alleviate logical inconsistencies and hallucinations in LLMs' problem-solving processes. However, we argue that the root challenge is more profound: LLMs lack the complex thinking paradigms (\ie, computational thinking) during reasoning. In this paper, we propose Computational Thinking Model (CTM), a novel framework that incorporates computational thinking paradigms into LLMs. This framework enables LLMs to reformulate complex problems through decomposition, abstraction, reduction, and simulation, among other techniques. Specifically, live code execution is seamlessly integrated into the reasoning process, allowing CTM to think by computing. CTM directly instills computational thinking objectives into LLMs through tailored reinforcement learning rewards, which encourages problem simplification, modular planning, and iterative verification. We conduct extensive evaluations on multiple code generation and mathematical benchmarks. The results demonstrate that CTM outperforms conventional reasoning models and tool-augmented baselines in terms of accuracy, interpretability, and generalizability. We hope this study offers valuable insights for AI reasoning, where LLMs can transform problems into robust, verifiable, and scalable computational workflows, much like computer scientists do. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:11:15 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:01:09 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zhang",
"Kechi",
""
],
[
"Li",
"Ge",
""
],
[
"Li",
"Jia",
""
],
[
"Zhang",
"Huangzhao",
""
],
[
"Xu",
"Jingjing",
""
],
[
"Zhu",
"Hao",
""
],
[
"Wang",
"Lecheng",
""
],
[
"Li",
"Jia",
""
],
[
"Dong",
"Yihong",
""
],
[
"Mai",
"Jing",
""
],
[
"Gu",
"Bin",
""
],
[
"Jin",
"Zhi",
""
]
] |
2506.02687 | Xiaowen Zhang | Hongxi Liu, Long-Tu Yuan, Xiaowen Zhang | A Fan-type condition involving bipartite independence number for hamiltonicity in graphs | We wish to withdraw the paper titled "A Fan-type condition involving bipartite independence number for hamiltonicity in graphs" due to an error in Section 2. A revised version of the paper will be submitted once the appropriate corrections have been made | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The bipartite independence number of a graph $G$, denoted by $\widetilde{\alpha}(G)$, is defined as the smallest integer $q$ for which there exist positive integers $s$ and $t$ satisfying $s + t = q + 1$, such that for any two disjoint subsets $A, B \subseteq V(G)$ with $|A| = s$ and $|B| = t$, there exists an edge between $A$ and $B$. In this paper, we prove that for a $2$-connected graph $G$ of order at least three, if $\max\{d_G(x),d_G(y)\}\geq \widetilde{\alpha}(G)$ for any pair of nonadjacent vertices $x$ and $y$ in $G$ at distance 2, then $G$ is hamiltonian. In addition, we prove that for a $3$-connected graph $G$, if $\max\{d_G(x),d_G(y)\}\geq \widetilde{\alpha}(G)+1$ for any pair of nonadjacent vertices $x$ and $y$ in $G$ at distance 2, then $G$ is hamiltonian-connected. Our results generalize the work of Li and Liu. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:40:28 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 09:45:23 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Liu",
"Hongxi",
""
],
[
"Yuan",
"Long-Tu",
""
],
[
"Zhang",
"Xiaowen",
""
]
] |
2506.02689 | Yihong Tang | Liang Yue, Yihong Tang, Kehai Chen, Jie Liu, Min Zhang | MASTER: Enhancing Large Language Model via Multi-Agent Simulated Teaching | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Instruction fine-tuning is crucial in NLP tasks, enhancing pretrained models' instruction-following capabilities and task-specific performance. However, obtaining high-quality fine-tuning data for large models is challenging due to data collection difficulties and high production costs. To address this, we propose MASTER, a novel data augmentation method that enriches original data through interactions among multiple agents with varying cognitive levels. We simulate three pedagogically grounded teaching scenarios, leveraging multi-agent conversations to generate high-quality teacher-student interaction data. Utilizing MASTER, we construct BOOST-QA, a fine-tuning dataset augmented from existing datasets like Orca-Math-200k, ProcQA, and OpenHermes2.5. Experiments show that models fine-tuned with BOOST-QA perform excellently across multiple benchmarks, demonstrating strong multitask generalization. Notably, MASTER significantly improves models' reasoning abilities in complex tasks, providing valuable insights for future research. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:41:35 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:34:54 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yue",
"Liang",
""
],
[
"Tang",
"Yihong",
""
],
[
"Chen",
"Kehai",
""
],
[
"Liu",
"Jie",
""
],
[
"Zhang",
"Min",
""
]
] |
2506.02695 | Linquan Wu | Linquan Wu, Tianxiang Jiang, Wenhao Duan, Yini Fang, Jacky Keung | FaceSleuth: Learning-Driven Single-Orientation Attention Verifies Vertical Dominance in Micro-Expression Recognition | 12 pages, 2 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Micro-expression recognition (MER) demands models that can amplify millisecond-level, low-amplitude facial motions while suppressing identity-specific appearance. We introduce FaceSleuth, a dual-stream architecture that (1) enhances motion along the empirically dominant vertical axix through a Continuously Vertical Attention (CVA) block, (2) localises the resulting signals with a Facial Position Focalizer built on hierarchical cross-window attention, and (3) steers feature learning toward physiologically meaningful regions via lightweight Action-Unit embeddings. To examine whether the hand-chosen vertical axis is indeed optimal, we further propose a Single-Orientation Attention (SOA) module that learns its own pooling direction end-to-end. SOA is differentiable, adds only 0.16 % parameters, and collapses to CVA when the learned angle converges to {\Pi}/2. In practice, SOA reliably drifts to 88{\deg}, confirming the effectiveness of the vertical prior while delivering consistent gains. On three standard MER benchmarks, FaceSleuth with CVA already surpasses previous state-of-the-art methods; plugging in SOA lifts accuracy and F1 score performance to 95.1 % / 0.918 on CASME II, 87.1 % / 0.840 on SAMM, and 92.9 % / 0.917 on MMEW without sacrificing model compactness. These results establish a new state of the art and, for the first time, provide empirical evidence that the vertical attention bias is the most discriminative orientation for MER. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:44:18 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 01:48:39 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wu",
"Linquan",
""
],
[
"Jiang",
"Tianxiang",
""
],
[
"Duan",
"Wenhao",
""
],
[
"Fang",
"Yini",
""
],
[
"Keung",
"Jacky",
""
]
] |
2506.02719 | Uwe R. Fischer | S. Mahesh Chandran and Uwe R. Fischer | Expansion-contraction duality breaking in a Planck-scale sensitive cosmological quantum simulator | 7+4 pages, 4+1 figures; authors whose pertinent work we inadvertently omitted from the citations, please e-mail us | null | null | null | gr-qc astro-ph.CO cond-mat.quant-gas quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose the experimental simulation of cosmological perturbations governed by a Planck-scale induced Lorentz violating dispersion, aimed at distinguishing between early-universe models with similar power spectra. Employing a novel variant of the scaling approach for the evolution of a Bose-Einstein condensate with both contact and dipolar interactions, we show that scale invariance, and in turn, the duality of the power spectrum is broken at large momenta for an inflating gas, and at small momenta for a contracting gas. We thereby furnish a Planck-scale sensitive approach to analogue quantum cosmology that can readily be implemented in the quantum gas laboratory. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:18:16 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chandran",
"S. Mahesh",
""
],
[
"Fischer",
"Uwe R.",
""
]
] |
2506.02731 | Joaquin Masias | Luis A. Anchordoqui, Ignatios Antoniadis, Niccol\`o Cribiori, Arda Hasar, Dieter Lust, Joaquin Masias, Marco Scalisi | Bulk/boundary Modular Quintessence and DESI | 21 pages, 4 figures | null | null | null | hep-th astro-ph.CO gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The latest DESI DR2 results, when combined with other independent cosmological data on the Cosmic Microwave Background and supernovas, suggest a preference for dynamical dark energy.
We propose a novel cosmological scenario, which features two distinct scalar fields. One governs the magnitude of the present-day dark energy density and is related to the size of extra-dimensions. Accounting for the observed smallness of this energy density requires the scalar to reside near the boundary of field space.
The second field, responsible for the time evolution of dark energy and associated with the string coupling, must instead lie in the bulk to remain consistent with the non-observation of light string states. We show that a natural candidate for such dark energy dynamics is a quintessence modular-invariant potential, in which the second scalar field rolls down a negatively curved slope, starting from a self-dual critical point. We find that this scenario is in good agreement with the latest findings by DESI. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:47:03 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Anchordoqui",
"Luis A.",
""
],
[
"Antoniadis",
"Ignatios",
""
],
[
"Cribiori",
"Niccolò",
""
],
[
"Hasar",
"Arda",
""
],
[
"Lust",
"Dieter",
""
],
[
"Masias",
"Joaquin",
""
],
[
"Scalisi",
"Marco",
""
]
] |
2506.02738 | Negin Baghbanzadeh | Negin Baghbanzadeh, Sajad Ashkezari, Elham Dolatabadi, Arash Afkanpour | Open-PMC-18M: A High-Fidelity Large Scale Medical Dataset for Multimodal Representation Learning | 15 pages | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Compound figures, which are multi-panel composites containing diverse subfigures, are ubiquitous in biomedical literature, yet large-scale subfigure extraction remains largely unaddressed. Prior work on subfigure extraction has been limited in both dataset size and generalizability, leaving a critical open question: How does high-fidelity image-text alignment via large-scale subfigure extraction impact representation learning in vision-language models? We address this gap by introducing a scalable subfigure extraction pipeline based on transformer-based object detection, trained on a synthetic corpus of 500,000 compound figures, and achieving state-of-the-art performance on both ImageCLEF 2016 and synthetic benchmarks. Using this pipeline, we release OPEN-PMC-18M, a large-scale high quality biomedical vision-language dataset comprising 18 million clinically relevant subfigure-caption pairs spanning radiology, microscopy, and visible light photography. We train and evaluate vision-language models on our curated datasets and show improved performance across retrieval, zero-shot classification, and robustness benchmarks, outperforming existing baselines. We release our dataset, models, and code to support reproducible benchmarks and further study into biomedical vision-language modeling and representation learning. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:53:19 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 12:14:31 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Baghbanzadeh",
"Negin",
""
],
[
"Ashkezari",
"Sajad",
""
],
[
"Dolatabadi",
"Elham",
""
],
[
"Afkanpour",
"Arash",
""
]
] |
2506.02774 | Yu Feng | Zheng Liu, He Zhu, Xinyang Li, Yirun Wang, Yujiao Shi, Wei Li, Jingwen Leng, Minyi Guo, Yu Feng | Voyager: Real-Time Splatting City-Scale 3D Gaussians on Your Phone | null | null | null | null | cs.GR | http://creativecommons.org/licenses/by/4.0/ | 3D Gaussian Splatting (3DGS) is an emerging technique for photorealistic 3D scene rendering. However, rendering city-scale 3DGS scenes on mobile devices, e.g., your smartphones, remains a significant challenge due to the limited resources on mobile devices. A natural solution is to offload computation to the cloud; however, naively streaming rendered frames from the cloud to the client introduces high latency and requires bandwidth far beyond the capacity of current wireless networks.
In this paper, we propose an effective solution to enable city-scale 3DGS rendering on mobile devices. Our key insight is that, under normal user motion, the number of newly visible Gaussians per second remains roughly constant. Leveraging this, we stream only the necessary Gaussians to the client. Specifically, on the cloud side, we propose asynchronous level-of-detail search to identify the necessary Gaussians for the client. On the client side, we accelerate rendering via a lookup table-based rasterization. Combined with holistic runtime optimizations, our system can deliver low-latency, city-scale 3DGS rendering on mobile devices. Compared to existing solutions, Voyager achieves over 100$\times$ reduction on data transfer and up to 8.9$\times$ speedup while retaining comparable rendering quality. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:50:51 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 01:40:43 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Liu",
"Zheng",
""
],
[
"Zhu",
"He",
""
],
[
"Li",
"Xinyang",
""
],
[
"Wang",
"Yirun",
""
],
[
"Shi",
"Yujiao",
""
],
[
"Li",
"Wei",
""
],
[
"Leng",
"Jingwen",
""
],
[
"Guo",
"Minyi",
""
],
[
"Feng",
"Yu",
""
]
] |
2506.02814 | Jinhao Sheng | Jinhao Sheng, Zhiqing Tang, Jianxiong Guo, Tian Wang | Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing | This paper is accepted by IEEE HPCC 2024 | null | 10.1109/HPCC64274.2024.00101 | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The growing demand for real-time processing tasks is driving the need for multi-model inference pipelines on edge devices. However, cost-effectively deploying these pipelines while optimizing Quality of Service (QoS) and costs poses significant challenges. Existing solutions often neglect device resource constraints, focusing mainly on inference accuracy and cost efficiency. To address this, we develop a framework for configuring multi-model inference pipelines. Specifically: 1) We model the decision-making problem by considering the pipeline's QoS, costs, and device resource limitations. 2) We create a feature extraction module using residual networks and a load prediction model based on Long Short-Term Memory (LSTM) to gather comprehensive node and pipeline status information. Then, we implement a Reinforcement Learning (RL) algorithm based on policy gradients for online configuration decisions. 3) Experiments conducted in a real Kubernetes cluster show that our approach significantly improve QoS while reducing costs and shorten decision-making time for complex pipelines compared to baseline algorithms. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:44:46 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 07:58:37 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Sheng",
"Jinhao",
""
],
[
"Tang",
"Zhiqing",
""
],
[
"Guo",
"Jianxiong",
""
],
[
"Wang",
"Tian",
""
]
] |
2506.02834 | Tin Tran | Tin T. Tran, Vaclav Snasel and Loc Tan Nguyen | Combining social relations and interaction data in Recommender System with Graph Convolution Collaborative Filtering | null | null | 10.1109/ACCESS.2023.3340209 | null | cs.IR | http://creativecommons.org/licenses/by-sa/4.0/ | A recommender system is an important subject in the field of data mining, where the item rating information from users is exploited and processed to make suitable recommendations with all other users. The recommender system creates convenience for e-commerce users and stimulates the consumption of items that are suitable for users. In addition to e-commerce, a recommender system is also used to provide recommendations on books to read, movies to watch, courses to take or websites to visit. Similarity between users is an important impact for recommendation, which could be calculated from the data of past user ratings of the item by methods of collaborative filtering, matrix factorization or singular vector decomposition. In the development of graph data mining techniques, the relationships between users and items can be represented by matrices from which collaborative filtering could be done with the larger database, more accurate and faster in calculation. All these data can be represented graphically and mined by today's highly developed graph neural network models. On the other hand, users' social friendship data also influence consumption habits because recommendations from friends will be considered more carefully than information sources. However, combining a user's friend influence and the similarity between users whose similar shopping habits is challenging. Because the information is noisy and it affects each particular data set in different ways. In this study, we present the input data processing method to remove outliers which are single reviews or users with little interaction with the items; the next proposed model will combine the social relationship data and the similarity in the rating history of users to improve the accuracy and recall of the recommender system. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 13:04:00 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Tran",
"Tin T.",
""
],
[
"Snasel",
"Vaclav",
""
],
[
"Nguyen",
"Loc Tan",
""
]
] |
2506.02845 | Di Wen | Di Wen, Lei Qi, Kunyu Peng, Kailun Yang, Fei Teng, Ao Luo, Jia Fu, Yufan Chen, Ruiping Liu, Yitian Shi, M. Saquib Sarfraz, Rainer Stiefelhagen | Go Beyond Earth: Understanding Human Actions and Scenes in Microgravity Environments | 15 pages, 3 figures, code are available at https://github.com/LEI-QI-233/HAR-in-Space | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Despite substantial progress in video understanding, most existing datasets are limited to Earth's gravitational conditions. However, microgravity alters human motion, interactions, and visual semantics, revealing a critical gap for real-world vision systems. This presents a challenge for domain-robust video understanding in safety-critical space applications. To address this, we introduce MicroG-4M, the first benchmark for spatio-temporal and semantic understanding of human activities in microgravity. Constructed from real-world space missions and cinematic simulations, the dataset includes 4,759 clips covering 50 actions, 1,238 context-rich captions, and over 7,000 question-answer pairs on astronaut activities and scene understanding. MicroG-4M supports three core tasks: fine-grained multi-label action recognition, temporal video captioning, and visual question answering, enabling a comprehensive evaluation of both spatial localization and semantic reasoning in microgravity contexts. We establish baselines using state-of-the-art models. All data, annotations, and code are available at https://github.com/LEI-QI-233/HAR-in-Space. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 13:15:19 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 14:43:51 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wen",
"Di",
""
],
[
"Qi",
"Lei",
""
],
[
"Peng",
"Kunyu",
""
],
[
"Yang",
"Kailun",
""
],
[
"Teng",
"Fei",
""
],
[
"Luo",
"Ao",
""
],
[
"Fu",
"Jia",
""
],
[
"Chen",
"Yufan",
""
],
[
"Liu",
"Ruiping",
""
],
[
"Shi",
"Yitian",
""
],
[
"Sarfraz",
"M. Saquib",
""
],
[
"Stiefelhagen",
"Rainer",
""
]
] |
2506.02867 | Chen Qian | Chen Qian, Dongrui Liu, Haochen Wen, Zhen Bai, Yong Liu, Jing Shao | Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning | Preprint. Under review | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large reasoning models (LRMs) have demonstrated impressive capabilities in complex problem-solving, yet their internal reasoning mechanisms remain poorly understood. In this paper, we investigate the reasoning trajectories of LRMs from an information-theoretic perspective. By tracking how mutual information (MI) between intermediate representations and the correct answer evolves during LRM reasoning, we observe an interesting MI peaks phenomenon: the MI at specific generative steps exhibits a sudden and significant increase during LRM's reasoning process. We theoretically analyze such phenomenon and show that as MI increases, the probability of model's prediction error decreases. Furthermore, these MI peaks often correspond to tokens expressing reflection or transition, such as ``Hmm'', ``Wait'' and ``Therefore,'' which we term as the thinking tokens. We then demonstrate that these thinking tokens are crucial for LRM's reasoning performance, while other tokens has minimal impacts. Building on these analyses, we propose two simple yet effective methods to improve LRM's reasoning performance, by delicately leveraging these thinking tokens. Overall, our work provides novel insights into the reasoning mechanisms of LRMs and offers practical ways to improve their reasoning capabilities. The code is available at https://github.com/ChnQ/MI-Peaks. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 13:31:10 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 15:00:58 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Qian",
"Chen",
""
],
[
"Liu",
"Dongrui",
""
],
[
"Wen",
"Haochen",
""
],
[
"Bai",
"Zhen",
""
],
[
"Liu",
"Yong",
""
],
[
"Shao",
"Jing",
""
]
] |
2506.02880 | Jiang Hu Dr. | Jian Cui, Jiang Hu, Zhidong Bai and Guorong Hu | On the rate of convergence in the CLT for LSS of large-dimensional sample covariance matrices | Modified some typos and added some references | null | null | null | math.PR math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper investigates the rate of convergence for the central limit theorem of linear spectral statistic (LSS) associated with large-dimensional sample covariance matrices. We consider matrices of the form ${\mathbf B}_n=\frac{1}{n}{\mathbf T}_p^{1/2}{\mathbf X}_n{\mathbf X}_n^*{\mathbf T}_p^{1/2},$ where ${\mathbf X}_n= (x_{i j} ) $ is a $p \times n$ matrix whose entries are independent and identically distributed (i.i.d.) real or complex variables, and ${\mathbf T} _p$ is a $p\times p$ nonrandom Hermitian nonnegative definite matrix with its spectral norm uniformly bounded in $p$. Employing Stein's method, we establish that if the entries $x_{ij}$ satisfy $\mathbb{E}|x_{ij}|^{10}<\infty$ and the ratio of the dimension to sample size $p/n\to y>0$ as $n\to\infty$, then the convergence rate of the normalized LSS of ${\mathbf B}_n$ to the standard normal distribution, measured in the Kolmogorov-Smirnov distance, is $O(n^{-1/2+\kappa})$ for any fixed $\kappa>0$. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 13:46:36 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 15:37:01 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Cui",
"Jian",
""
],
[
"Hu",
"Jiang",
""
],
[
"Bai",
"Zhidong",
""
],
[
"Hu",
"Guorong",
""
]
] |
2506.02896 | Adam Pardyl | Adam Pardyl, Dominik Matuszek, Mateusz Przebieracz, Marek Cygan, Bartosz Zieli\'nski, Maciej Wo{\l}czyk | FlySearch: Exploring how vision-language models explore | null | null | null | null | cs.CV cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The real world is messy and unstructured. Uncovering critical information often requires active, goal-driven exploration. It remains to be seen whether Vision-Language Models (VLMs), which recently emerged as a popular zero-shot tool in many difficult tasks, can operate effectively in such conditions. In this paper, we answer this question by introducing FlySearch, a 3D, outdoor, photorealistic environment for searching and navigating to objects in complex scenes. We define three sets of scenarios with varying difficulty and observe that state-of-the-art VLMs cannot reliably solve even the simplest exploration tasks, with the gap to human performance increasing as the tasks get harder. We identify a set of central causes, ranging from vision hallucination, through context misunderstanding, to task planning failures, and we show that some of them can be addressed by finetuning. We publicly release the benchmark, scenarios, and the underlying codebase. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 14:03:42 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 09:32:09 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Pardyl",
"Adam",
""
],
[
"Matuszek",
"Dominik",
""
],
[
"Przebieracz",
"Mateusz",
""
],
[
"Cygan",
"Marek",
""
],
[
"Zieliński",
"Bartosz",
""
],
[
"Wołczyk",
"Maciej",
""
]
] |
2506.02916 | Hao Fan | Hao Fan, Yanrong Hu, Kai Fang, Qingyang Liu, Hongjiu Liu | MMM4Rec: A Transfer-Efficient Framework for Multi-modal Sequential Recommendation | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sequential Recommendation (SR) systems model user preferences by analyzing interaction histories. Although transferable multi-modal SR architectures demonstrate superior performance compared to traditional ID-based approaches, current methods incur substantial fine-tuning costs when adapting to new domains due to complex optimization requirements and negative transfer effects - a significant deployment bottleneck that hinders engineers from efficiently repurposing pre-trained models for novel application scenarios with minimal tuning overhead. We propose MMM4Rec (Multi-Modal Mamba for Sequential Recommendation), a novel multi-modal SR framework that incorporates a dedicated algebraic constraint mechanism for efficient transfer learning. By combining State Space Duality (SSD)'s temporal decay properties with a time-aware modeling design, our model dynamically prioritizes key modality information, overcoming limitations of Transformer-based approaches. The framework implements a constrained two-stage process: (1) sequence-level cross-modal alignment via shared projection matrices, followed by (2) temporal fusion using our newly designed Cross-SSD module and dual-channel Fourier adaptive filtering. This architecture maintains semantic consistency while suppressing noise propagation.MMM4Rec achieves rapid fine-tuning convergence with simple cross-entropy loss, significantly improving multi-modal recommendation accuracy while maintaining strong transferability. Extensive experiments demonstrate MMM4Rec's state-of-the-art performance, achieving the maximum 31.78% NDCG@10 improvement over existing models and exhibiting 10 times faster average convergence speed when transferring to large-scale downstream datasets. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 14:18:19 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 01:42:22 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Fan",
"Hao",
""
],
[
"Hu",
"Yanrong",
""
],
[
"Fang",
"Kai",
""
],
[
"Liu",
"Qingyang",
""
],
[
"Liu",
"Hongjiu",
""
]
] |
2506.02923 | Alexis Bellot | Alexis Bellot, Jonathan Richens, Tom Everitt | The Limits of Predicting Agents from Behaviour | null | ICML 2025 | null | null | cs.AI stat.ML | http://creativecommons.org/licenses/by/4.0/ | As the complexity of AI systems and their interactions with the world increases, generating explanations for their behaviour is important for safely deploying AI. For agents, the most natural abstractions for predicting behaviour attribute beliefs, intentions and goals to the system. If an agent behaves as if it has a certain goal or belief, then we can make reasonable predictions about how it will behave in novel situations, including those where comprehensive safety evaluations are untenable. How well can we infer an agent's beliefs from their behaviour, and how reliably can these inferred beliefs predict the agent's behaviour in novel situations? We provide a precise answer to this question under the assumption that the agent's behaviour is guided by a world model. Our contribution is the derivation of novel bounds on the agent's behaviour in new (unseen) deployment environments, which represent a theoretical limit for predicting intentional agents from behavioural data alone. We discuss the implications of these results for several research areas including fairness and safety. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 14:24:58 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Bellot",
"Alexis",
""
],
[
"Richens",
"Jonathan",
""
],
[
"Everitt",
"Tom",
""
]
] |
2506.02943 | Qinghua Xu | Qinghua Xu, Guancheng Wang, Lionel Briand, Kui Liu | A Multi-agent LLM-based JUnit Test Generation with Strong Oracles | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unit testing plays a critical role in ensuring software correctness. However, writing unit tests manually is laborious, especially for strong typed languages like Java, motivating the need for automated approaches. Traditional methods primarily rely on search-based or randomized algorithms to generate tests that achieve high code coverage and produce regression oracles, which are derived from the program's current behavior rather than its intended functionality. Recent advances in large language models (LLMs) have enabled oracle generation from natural language descriptions. However, existing LLM-based methods often require LLM fine-tuning or rely on external tools such as EvoSuite for test prefix generation.
In this work, we propose CANDOR, a novel end-to-end, prompt-based LLM framework for automated JUnit test generation. CANDOR orchestrates multiple specialized LLM agents to generate JUnit tests, including both high-quality test prefixes and accurate oracles. To mitigate the notorious hallucinations in LLMs, we introduce a novel strategy that engages multiple reasoning LLMs in a panel discussion and generate accurate oracles based on consensus. Additionally, to reduce the verbosity of reasoning LLMs' outputs, we propose a novel dual-LLM pipeline to produce concise and structured oracle evaluations.
Our experiments on the HumanEvalJava and LeetCodeJava datasets show that CANDOR can generate accurate oracles and is slightly better than EvoSuite in generating tests with high line coverage and clearly superior in terms of mutation score. Moreover, CANDOR significantly outperforms the state-of-the-art, prompt-based test generator LLM-Empirical, achieving improvements of 15.8 to 25.1 percentage points in oracle correctness on both correct and faulty source code. Ablation studies confirm the critical contributions of key agents in improving test prefix quality and oracle accuracy. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 14:43:05 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 14:43:39 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Xu",
"Qinghua",
""
],
[
"Wang",
"Guancheng",
""
],
[
"Briand",
"Lionel",
""
],
[
"Liu",
"Kui",
""
]
] |
2506.02965 | Ze Yu Zhang | Ze Yu Zhang, Bolin Ding, Bryan Kian Hsiang Low | PC-MoE: Memory-Efficient and Privacy-Preserving Collaborative Training for Mixture-of-Experts LLMs | 20 pages, 4 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mixture-of-Experts (MoE) has been gaining popularity due to its successful adaptation to large language models (LLMs). In this work, we introduce Privacy-preserving Collaborative Mixture-of-Experts (PC-MoE), which leverages the sparsity of the MoE architecture for memory-efficient decentralized collaborative LLM training, enabling multiple parties with limited GPU-memory and data resources to collectively train more capable LLMs than they could achieve individually. At the same time, this approach protects training data privacy of each participant by keeping training data, as well as parts of the forward pass signal and gradients locally within each party. By design, PC-MoE synergistically combines the strengths of distributed computation with strong confidentiality assurances. Unlike most privacy-preserving schemes, which pay for confidentiality with lower task accuracy, our framework breaks that trade-off: across seven popular LLM benchmarks, it almost matches (and sometimes exceeds) the performance and convergence rate of a fully centralized model, enjoys near 70% peak GPU RAM reduction, while being fully robust against reconstruction attacks. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 15:00:18 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 05:38:31 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zhang",
"Ze Yu",
""
],
[
"Ding",
"Bolin",
""
],
[
"Low",
"Bryan Kian Hsiang",
""
]
] |
2506.02970 | Tian-Hua Yang | Tian-Hua Yang and Dmitry A. Abanin | Integrals of motion as slow modes in dissipative many-body operator dynamics | null | null | null | null | quant-ph cond-mat.quant-gas cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider Lindbladian operator dynamics in many-body quantum systems with one or more integrals of motion (IOM), subject to weak local dissipation. We demonstrate that IOMs with small support become slow modes of these dynamics, in the sense that their Frobenius norm decays more slowly compared to generic operators. As a result, the eigenoperators of such Lindbladians with slowest decay rates have a large overlap with the IOMs of the underlying Hamiltonian. We demonstrate this correspondence between slow modes and IOMs numerically for a number of many-body models, and further corroborate it with perturbative arguments. These results open up a new method for the identification of IOMs, and provide insights into the dissipative many-body dynamics. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 15:06:01 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 09:03:57 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yang",
"Tian-Hua",
""
],
[
"Abanin",
"Dmitry A.",
""
]
] |
2506.03023 | Anowar J. Shajib | TDCOSMO Collaboration: Simon Birrer, Elizabeth J. Buckley-Geer, Michele Cappellari, Fr\'ed\'eric Courbin, Fr\'ed\'eric Dux, Christopher D. Fassnacht, Joshua A. Frieman, Aymeric Galan, Daniel Gilman, Xiang-Yu Huang, Shawn Knabel, Danial Langeroodi, Huan Lin, Martin Millon, Takahiro Morishita, Veronica Motta, Pritom Mozumdar, Eric Paic, Anowar J. Shajib, William Sheu, Dominique Sluse, Alessandro Sonnenfeld, Chiara Spiniello, Massimo Stiavelli, Sherry H. Suyu, Chin Yi Tan, Tommaso Treu, Lyne Van de Vyvere, Han Wang, Patrick Wells, Devon M. Williams, and Kenneth C. Wong | TDCOSMO 2025: Cosmological constraints from strong lensing time delays | 34 pages, 17 figures, 8 tables, (this version: fixed author names). A virtual talk on the presented results will be given in the CosmoVerse Seminar on June 12, 2025, at 4pm UK time, details: https://cosmoversetensions.eu/for-scientists/cosmoverse-seminars/ | null | null | null | astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | We present cosmological constraints from 8 strongly lensed quasars (hereafter, the TDCOSMO-2025 sample). Building on previous work, our analysis incorporated new deflector stellar velocity dispersions measured from spectra obtained with the James Webb Space Telescope (JWST), the Keck Telescopes, and the Very Large Telescope (VLT), utilizing improved methods. We used integrated JWST stellar kinematics for 5 lenses, VLT-MUSE for 2, and resolved kinematics from Keck and JWST for RX J1131-1231. We also considered two samples of non-time-delay lenses: 11 from the Sloan Lens ACS (SLACS) sample with Keck-KCWI resolved kinematics; and 4 from the Strong Lenses in the Legacy Survey (SL2S) sample. We improved our analysis of line-of-sight effects, the surface brightness profile of the lens galaxies, and orbital anisotropy, and corrected for projection effects in the dynamics. Our uncertainties are maximally conservative by accounting for the mass-sheet degeneracy in the deflectors' mass density profiles. The analysis was blinded to prevent experimenter bias. Our primary result is based on the TDCOSMO-2025 sample, in combination with $\Omega_{\rm m}$ constraints from the Pantheon+ Type Ia supernovae (SN) dataset. In the flat $\Lambda$ Cold Dark Matter (CDM), we find $H_0=72.1^{+4.0}_{-3.7}$ km s$^{-1}$ Mpc$^{-1}$. The SLACS and SL2S samples are in excellent agreement with the TDCOSMO-2025 sample, improving the precision on $H_0$ in flat $\Lambda$CDM to 4.4%. Using the Dark Energy Survey SN Year-5 dataset (DES-SN5YR) or DESI-DR2 baryonic acoustic oscillations (BAO) likelihoods instead of Pantheon+ yields very similar results. We also present constraints in the open $\Lambda$CDM, $w$CDM, $w_0w_a$CDM, and $w_{\phi}$CDM cosmologies. The TDCOSMO $H_0$ inference is robust and consistent across all presented cosmological models, and our cosmological constraints in them agree with those from the BAO and SN. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 15:59:29 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 17:22:25 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"TDCOSMO Collaboration",
"",
""
],
[
"Birrer",
"Simon",
""
],
[
"Buckley-Geer",
"Elizabeth J.",
""
],
[
"Cappellari",
"Michele",
""
],
[
"Courbin",
"Frédéric",
""
],
[
"Dux",
"Frédéric",
""
],
[
"Fassnacht",
"Christopher D.",
""
],
[
"Frieman",
"Joshua A.",
""
],
[
"Galan",
"Aymeric",
""
],
[
"Gilman",
"Daniel",
""
],
[
"Huang",
"Xiang-Yu",
""
],
[
"Knabel",
"Shawn",
""
],
[
"Langeroodi",
"Danial",
""
],
[
"Lin",
"Huan",
""
],
[
"Millon",
"Martin",
""
],
[
"Morishita",
"Takahiro",
""
],
[
"Motta",
"Veronica",
""
],
[
"Mozumdar",
"Pritom",
""
],
[
"Paic",
"Eric",
""
],
[
"Shajib",
"Anowar J.",
""
],
[
"Sheu",
"William",
""
],
[
"Sluse",
"Dominique",
""
],
[
"Sonnenfeld",
"Alessandro",
""
],
[
"Spiniello",
"Chiara",
""
],
[
"Stiavelli",
"Massimo",
""
],
[
"Suyu",
"Sherry H.",
""
],
[
"Tan",
"Chin Yi",
""
],
[
"Treu",
"Tommaso",
""
],
[
"Van de Vyvere",
"Lyne",
""
],
[
"Wang",
"Han",
""
],
[
"Wells",
"Patrick",
""
],
[
"Williams",
"Devon M.",
""
],
[
"Wong",
"Kenneth C.",
""
]
] |
2506.03045 | Lucas Emanuel Antunes Porto | Lucas E. A. Porto, S\'ebastien Designolle, Sebastian Pokutta and Marco T\'ulio Quintino | Measurement incompatibility and quantum steering via linear programming | null | null | null | null | quant-ph math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The problem of deciding whether a set of quantum measurements is jointly measurable is known to be equivalent to determining whether a quantum assemblage is unsteerable. This problem can be formulated as a semidefinite program (SDP). However, the number of variables and constraints in such a formulation grows exponentially with the number of measurements, rendering it intractable for large measurement sets. In this work, we circumvent this problem by transforming the SDP into a hierarchy of linear programs that compute upper and lower bounds on the incompatibility robustness with a complexity that grows polynomially in the number of measurements. The hierarchy is guaranteed to converge and it can be applied to arbitrary measurements -- including non-projective POVMs -- in arbitrary dimensions. While convergence becomes impractical in high dimensions, in the case of qubits our method reliably provides accurate upper and lower bounds for the incompatibility robustness of sets with several hundred measurements in a short time using a standard laptop. We also apply our methods to qutrits, obtaining non-trivial upper and lower bounds in scenarios that are otherwise intractable using the standard SDP approach. Finally, we show how our methods can be used to construct local hidden state models for states, or conversely, to certify that a given state exhibits steering; for two-qubit quantum states, our approach is comparable to, and in some cases outperforms, the current best methods. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 16:28:25 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Porto",
"Lucas E. A.",
""
],
[
"Designolle",
"Sébastien",
""
],
[
"Pokutta",
"Sebastian",
""
],
[
"Quintino",
"Marco Túlio",
""
]
] |
2506.03074 | Junghyun Lee | Junghyun Lee and Kyoungseok Jang and Kwang-Sung Jun and Milan Vojnovi\'c and Se-Young Yun | GL-LowPopArt: A Nearly Instance-Wise Minimax-Optimal Estimator for Generalized Low-Rank Trace Regression | 53 pages, 2 figures, 3 tables; Accepted as a Spotlight Poster to the 42nd International Conference on Machine Learning (ICML 2025). Minor correction to the arXiv title in v2 ;) | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | We present `GL-LowPopArt`, a novel Catoni-style estimator for generalized low-rank trace regression. Building on `LowPopArt` (Jang et al., 2024), it employs a two-stage approach: nuclear norm regularization followed by matrix Catoni estimation. We establish state-of-the-art estimation error bounds, surpassing existing guarantees (Fan et al., 2019; Kang et al., 2022), and reveal a novel experimental design objective, $\mathrm{GL}(\pi)$. The key technical challenge is controlling bias from the nonlinear inverse link function, which we address by our two-stage approach. We prove a *local* minimax lower bound, showing that our `GL-LowPopArt` enjoys instance-wise optimality up to the condition number of the ground-truth Hessian. Applications include generalized linear matrix completion, where `GL-LowPopArt` achieves a state-of-the-art Frobenius error guarantee, and **bilinear dueling bandits**, a novel setting inspired by general preference learning (Zhang et al., 2024). Our analysis of a `GL-LowPopArt`-based explore-then-commit algorithm reveals a new, potentially interesting problem-dependent quantity, along with improved Borda regret bound than vectorization (Wu et al., 2024). | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 16:52:24 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 02:09:24 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Lee",
"Junghyun",
""
],
[
"Jang",
"Kyoungseok",
""
],
[
"Jun",
"Kwang-Sung",
""
],
[
"Vojnović",
"Milan",
""
],
[
"Yun",
"Se-Young",
""
]
] |
2506.03106 | Xiaoying Zhang | Xiaoying Zhang, Hao Sun, Yipeng Zhang, Kaituo Feng, Chaochao Lu, Chao Yang, Helen Meng | Critique-GRPO: Advancing LLM Reasoning with Natural Language and Numerical Feedback | 38 pages | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in reinforcement learning (RL) with numerical feedback, such as scalar rewards, have significantly enhanced the complex reasoning capabilities of large language models (LLMs). Despite this success, we identify three key challenges encountered by RL with solely numerical feedback: performance plateaus, limited effectiveness of self-reflection, and persistent failures. We then demonstrate that RL-finetuned models, even after exhibiting performance plateaus, can generate correct refinements on persistently failed problems by leveraging natural language feedback in the form of critiques. Building on this insight, we propose Critique-GRPO, an online RL framework that integrates both natural language and numerical feedback for effective policy optimization. Critique-GRPO enables LLMs to learn from initial responses and critique-guided refinements simultaneously while maintaining exploration. Extensive experiments using Qwen2.5-7B-Base and Qwen3-8B-Base show that Critique-GRPO consistently outperforms supervised learning-based and RL-based fine-tuning approaches across eight challenging mathematical, STEM, and general reasoning tasks, improving average pass@1 scores by approximately 4.5% and 5%, respectively. Notably, Critique-GRPO surpasses a strong baseline that incorporates expert demonstrations within online RL. Further analysis reveals two critical insights about policy exploration: (1) higher entropy does not always guarantee efficient learning from exploration, and (2) longer responses do not necessarily lead to more effective exploration. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 17:39:02 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 13:45:47 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zhang",
"Xiaoying",
""
],
[
"Sun",
"Hao",
""
],
[
"Zhang",
"Yipeng",
""
],
[
"Feng",
"Kaituo",
""
],
[
"Lu",
"Chaochao",
""
],
[
"Yang",
"Chao",
""
],
[
"Meng",
"Helen",
""
]
] |
2506.03151 | Ruibo Wang | Ruibo Wang, Mustafa A. Kishk, Howard H. Yang, and Mohamed-Slim Alouini | Analyzing Localizability of LEO/MEO Hybrid Networks: A Stochastic Geometry Approach | null | null | null | null | cs.NI | http://creativecommons.org/publicdomain/zero/1.0/ | With the increase in global positioning service demands and the requirement for more precise positioning, assisting existing medium and high orbit satellite-enabled positioning systems with low Earth orbit (LEO) satellites has garnered widespread attention. However, providing low computational complexity performance analysis for hybrid LEO/MEO massive satellite constellations remains a challenge. In this article, we introduce for the first time the application of stochastic geometry (SG) framework in satellite-enabled positioning performance analysis and provide an analytical expression for the K-availiability probability and K-localizability probability under bidirectional beam alignment transmissions. The K-localizability probability, defined as the probability that at least K satellites can participate in the positioning process, serves as a prerequisite for positioning. Since the modeling of MEO satellite constellations within the SG framework has not yet been studied, we integrate the advantages of Cox point processes and binomial point processes, proposing a doubly stochastic binomial point process binomial point process for accurate modeling of MEO satellite constellations. Finally, we investigate the impact of constellation configurations and antenna patterns on the localizability performance of LEO, MEO, and hybrid MEO/LEO constellations. We also demonstrate the network performance gains brought to MEO positioning systems by incorporating assistance from LEO satellites. | [
{
"version": "v1",
"created": "Fri, 11 Apr 2025 21:57:03 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wang",
"Ruibo",
""
],
[
"Kishk",
"Mustafa A.",
""
],
[
"Yang",
"Howard H.",
""
],
[
"Alouini",
"Mohamed-Slim",
""
]
] |
2506.03152 | Robert Bayer | Robert Bayer, Julian Priest, Daniel Kjellberg, Jeppe Lindhard, Nikolaj S{\o}renesen, Nicolaj Valsted, \'Ivar \'Oli, P{\i}nar T\"oz\"un | Adaptive and Robust Image Processing on CubeSats | null | null | null | null | eess.IV cs.CV cs.DC cs.LG | http://creativecommons.org/licenses/by/4.0/ | CubeSats offer a low-cost platform for space research, particularly for Earth observation. However, their resource-constrained nature and being in space, challenge the flexibility and complexity of the deployed image processing pipelines and their orchestration. This paper introduces two novel systems, DIPP and DISH, to address these challenges. DIPP is a modular and configurable image processing pipeline framework that allows for adaptability to changing mission goals even after deployment, while preserving robustness. DISH is a domain-specific language (DSL) and runtime system designed to schedule complex imaging workloads on low-power and memory-constrained processors.
Our experiments demonstrate that DIPP's decomposition of the processing pipelines adds negligible overhead, while significantly reducing the network requirements of updating pipelines and being robust against erroneous module uploads. Furthermore, we compare DISH to Lua, a general purpose scripting language, and demonstrate its comparable expressiveness and lower memory requirement. | [
{
"version": "v1",
"created": "Fri, 16 May 2025 16:03:04 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Bayer",
"Robert",
""
],
[
"Priest",
"Julian",
""
],
[
"Kjellberg",
"Daniel",
""
],
[
"Lindhard",
"Jeppe",
""
],
[
"Sørenesen",
"Nikolaj",
""
],
[
"Valsted",
"Nicolaj",
""
],
[
"Óli",
"Ívar",
""
],
[
"Tözün",
"Pınar",
""
]
] |
2506.03153 | Junzhe Jiang | Junzhe Jiang, Chang Yang, Xinrun Wang, Bo Li | Why Regression? Binary Encoding Classification Brings Confidence to Stock Market Index Price Prediction | null | null | null | null | q-fin.ST cs.LG | http://creativecommons.org/licenses/by/4.0/ | Stock market indices serve as fundamental market measurement that quantify systematic market dynamics. However, accurate index price prediction remains challenging, primarily because existing approaches treat indices as isolated time series and frame the prediction as a simple regression task. These methods fail to capture indices' inherent nature as aggregations of constituent stocks with complex, time-varying interdependencies. To address these limitations, we propose Cubic, a novel end-to-end framework that explicitly models the adaptive fusion of constituent stocks for index price prediction. Our main contributions are threefold. i) Fusion in the latent space: we introduce the fusion mechanism over the latent embedding of the stocks to extract the information from the vast number of stocks. ii) Binary encoding classification: since regression tasks are challenging due to continuous value estimation, we reformulate the regression into the classification task, where the target value is converted to binary and we optimize the prediction of the value of each digit with cross-entropy loss. iii) Confidence-guided prediction and trading: we introduce the regularization loss to address market prediction uncertainty for the index prediction and design the rule-based trading policies based on the confidence. Extensive experiments across multiple stock markets and indices demonstrate that Cubic consistently outperforms state-of-the-art baselines in stock index prediction tasks, achieving superior performance on both forecasting accuracy metrics and downstream trading profitability. | [
{
"version": "v1",
"created": "Sun, 18 May 2025 15:45:41 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Jiang",
"Junzhe",
""
],
[
"Yang",
"Chang",
""
],
[
"Wang",
"Xinrun",
""
],
[
"Li",
"Bo",
""
]
] |
2506.03154 | Zhaoyang Chen | Zhaoyang Chen, Cody Fleming | Modular Diffusion Policy Training: Decoupling and Recombining Guidance and Diffusion for Offline RL | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Classifier free guidance has shown strong potential in diffusion-based reinforcement learning. However, existing methods rely on joint training of the guidance module and the diffusion model, which can be suboptimal during the early stages when the guidance is inaccurate and provides noisy learning signals. In offline RL, guidance depends solely on offline data: observations, actions, and rewards, and is independent of the policy module's behavior, suggesting that joint training is not required. This paper proposes modular training methods that decouple the guidance module from the diffusion model, based on three key findings:
Guidance Necessity: We explore how the effectiveness of guidance varies with the training stage and algorithm choice, uncovering the roles of guidance and diffusion. A lack of good guidance in the early stage presents an opportunity for optimization.
Guidance-First Diffusion Training: We introduce a method where the guidance module is first trained independently as a value estimator, then frozen to guide the diffusion model using classifier-free reward guidance. This modularization reduces memory usage, improves computational efficiency, and enhances both sample efficiency and final performance.
Cross-Module Transferability: Applying two independently trained guidance models, one during training and the other during inference, can significantly reduce normalized score variance (e.g., reducing IQR by 86%). We show that guidance modules trained with one algorithm (e.g., IDQL) can be directly reused with another (e.g., DQL), with no additional training required, demonstrating baseline-level performance as well as strong modularity and transferability.
We provide theoretical justification and empirical validation on bullet D4RL benchmarks. Our findings suggest a new paradigm for offline RL: modular, reusable, and composable training pipelines. | [
{
"version": "v1",
"created": "Mon, 19 May 2025 22:51:58 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chen",
"Zhaoyang",
""
],
[
"Fleming",
"Cody",
""
]
] |
2506.03155 | Yu Zheng | Yu Zheng | Fusing Cross-Domain Knowledge from Multimodal Data to Solve Problems in the Physical World | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The proliferation of artificial intelligence has enabled a diversity of applications that bridge the gap between digital and physical worlds. As physical environments are too complex to model through a single information acquisition approach, it is crucial to fuse multimodal data generated by different sources, such as sensors, devices, systems, and people, to solve a problem in the real world. Unfortunately, it is neither applicable nor sustainable to deploy new resources to collect original data from scratch for every problem. Thus, when data is inadequate in the domain of problem, it is vital to fuse knowledge from multimodal data that is already available in other domains. We call this cross-domain knowledge fusion. Existing research focus on fusing multimodal data in a single domain, supposing the knowledge from different datasets is intrinsically aligned; however, this assumption may not hold in the scenarios of cross-domain knowledge fusion. In this paper, we formally define the cross-domain multimodal data fusion problem, discussing its unique challenges, differences and advantages beyond data fusion in a single domain. We propose a four-layer framework, consisting of Domains, Links, Models and Data layers, answering three key questions: "what to fuse", "why can be fused", and "how to fuse". The Domains Layer selects relevant data from different domains for a given problem. The Links Layer reveals the philosophy of knowledge alignment beyond specific model structures. The Models Layer provides two knowledge fusion paradigms based on the fundamental mechanisms for processing data. The Data Layer turns data of different structures, resolutions, scales and distributions into a consistent representation that can be fed into an AI model. With this framework, we can design end-to-end solutions that fuse cross-domain multimodal data effectively for solving real-world problems. | [
{
"version": "v1",
"created": "Tue, 20 May 2025 02:59:34 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zheng",
"Yu",
""
]
] |
2506.03156 | Davit Gondauri | Davit Gondauri | Gauging Growth: AGI Mathematical Metrics for Economic Progress | null | null | null | null | q-fin.GN | http://creativecommons.org/licenses/by/4.0/ | Today, the economy is greatly influenced by Artificial General Intelligence (AGI). The purpose of this paper is to determine the impact of the quantitative relations of AGI on the country's economic parameters. The authors use the analysis of historical data in the research, develop a new mathematical algorithm that refers to the level of AGI development, and conduct a regression analysis. The economic effect of AGI is deduced if it affects the growth of real GDP. As a result of the analysis, it is revealed that there is a positive Pearson correlation between the growth of AGI and real GDP; that is, to increase GDP by 1%, an average increase of 12.5% of AGI is required. | [
{
"version": "v1",
"created": "Tue, 20 May 2025 08:44:30 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Gondauri",
"Davit",
""
]
] |
2506.03158 | Jiahao Qin | Jiahao Qin, Bei Peng, Feng Liu, Guangliang Cheng, Lu Zong | DUAL: Dynamic Uncertainty-Aware Learning | 12 pages, 3 figures | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Deep learning models frequently encounter feature uncertainty in diverse learning scenarios, significantly impacting their performance and reliability. This challenge is particularly complex in multi-modal scenarios, where models must integrate information from different sources with inherent uncertainties. We propose Dynamic Uncertainty-Aware Learning (DUAL), a unified framework that effectively handles feature uncertainty in both single-modal and multi-modal scenarios. DUAL introduces three key innovations: Dynamic Feature Uncertainty Modeling, which continuously refines uncertainty estimates through joint consideration of feature characteristics and learning dynamics; Adaptive Distribution-Aware Modulation, which maintains balanced feature distributions through dynamic sample influence adjustment; and Uncertainty-aware Cross-Modal Relationship Learning, which explicitly models uncertainties in cross-modal interactions. Through extensive experiments, we demonstrate DUAL's effectiveness across multiple domains: in computer vision tasks, it achieves substantial improvements of 7.1% accuracy on CIFAR-10, 6.5% accuracy on CIFAR-100, and 2.3% accuracy on Tiny-ImageNet; in multi-modal learning, it demonstrates consistent gains of 4.1% accuracy on CMU-MOSEI and 2.8% accuracy on CMU-MOSI for sentiment analysis, while achieving 1.4% accuracy improvements on MISR. The code will be available on GitHub soon. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 18:50:15 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Qin",
"Jiahao",
""
],
[
"Peng",
"Bei",
""
],
[
"Liu",
"Feng",
""
],
[
"Cheng",
"Guangliang",
""
],
[
"Zong",
"Lu",
""
]
] |
2506.03159 | Lesley Wheat | Lesley Wheat, Martin v. Mohrenschildt and Saeid Habibi | Bayes Error Rate Estimation in Difficult Situations | 21 pages, 13 figures, 20 tables | null | null | null | cs.LG stat.ME stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Bayes Error Rate (BER) is the fundamental limit on the achievable generalizable classification accuracy of any machine learning model due to inherent uncertainty within the data. BER estimators offer insight into the difficulty of any classification problem and set expectations for optimal classification performance. In order to be useful, the estimators must also be accurate with a limited number of samples on multivariate problems with unknown class distributions. To determine which estimators meet the minimum requirements for "usefulness", an in-depth examination of their accuracy is conducted using Monte Carlo simulations with synthetic data in order to obtain their confidence bounds for binary classification. To examine the usability of the estimators on real-world applications, new test scenarios are introduced upon which 2500 Monte Carlo simulations per scenario are run over a wide range of BER values. In a comparison of k-Nearest Neighbor (kNN), Generalized Henze-Penrose (GHP) divergence and Kernel Density Estimation (KDE) techniques, results show that kNN is overwhelmingly the more accurate non-parametric estimator. In order to reach the target of an under 5 percent range for the 95 percent confidence bounds, the minimum number of required samples per class is 1000. As more features are added, more samples are needed, so that 2500 samples per class are required at only 4 features. Other estimators do become more accurate than kNN as more features are added, but continuously fail to meet the target range. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 19:58:20 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wheat",
"Lesley",
""
],
[
"Mohrenschildt",
"Martin v.",
""
],
[
"Habibi",
"Saeid",
""
]
] |
2506.03160 | Shriyank Somvanshi | Shriyank Somvanshi, Anannya Ghosh Tusti, Mahmuda Sultana Mimi, Md Monzurul Islam, Sazzad Bin Bashar Polock, Anandi Dutta, Subasish Das | Applying MambaAttention, TabPFN, and TabTransformers to Classify SAE Automation Levels in Crashes | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The increasing presence of automated vehicles (AVs) presents new challenges for crash classification and safety analysis. Accurately identifying the SAE automation level involved in each crash is essential to understanding crash dynamics and system accountability. However, existing approaches often overlook automation-specific factors and lack model sophistication to capture distinctions between different SAE levels. To address this gap, this study evaluates the performance of three advanced tabular deep learning models MambaAttention, TabPFN, and TabTransformer for classifying SAE automation levels using structured crash data from Texas (2024), covering 4,649 cases categorized as Assisted Driving (SAE Level 1), Partial Automation (SAE Level 2), and Advanced Automation (SAE Levels 3-5 combined). Following class balancing using SMOTEENN, the models were trained and evaluated on a unified dataset of 7,300 records. MambaAttention demonstrated the highest overall performance (F1-scores: 88% for SAE 1, 97% for SAE 2, and 99% for SAE 3-5), while TabPFN excelled in zero-shot inference with high robustness for rare crash categories. In contrast, TabTransformer underperformed, particularly in detecting Partial Automation crashes (F1-score: 55%), suggesting challenges in modeling shared human-system control dynamics. These results highlight the capability of deep learning models tailored for tabular data to enhance the accuracy and efficiency of automation-level classification. Integrating such models into crash analysis frameworks can support policy development, AV safety evaluation, and regulatory decisions, especially in distinguishing high-risk conditions for mid- and high-level automation technologies. | [
{
"version": "v1",
"created": "Thu, 22 May 2025 16:08:21 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Somvanshi",
"Shriyank",
""
],
[
"Tusti",
"Anannya Ghosh",
""
],
[
"Mimi",
"Mahmuda Sultana",
""
],
[
"Islam",
"Md Monzurul",
""
],
[
"Polock",
"Sazzad Bin Bashar",
""
],
[
"Dutta",
"Anandi",
""
],
[
"Das",
"Subasish",
""
]
] |
2506.03161 | Mira Nuthakki | Mira Nuthakki | Safety-Prioritized, Reinforcement Learning-Enabled Traffic Flow Optimization in a 3D City-Wide Simulation Environment | 18 pages, figures at end, methods at end. Format/order can be changed if necessary | null | null | null | cs.LG cs.AI cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traffic congestion and collisions represent significant economic, environmental, and social challenges worldwide. Traditional traffic management approaches have shown limited success in addressing these complex, dynamic problems. To address the current research gaps, three potential tools are developed: a comprehensive 3D city-wide simulation environment that integrates both macroscopic and microscopic traffic dynamics; a collision model; and a reinforcement learning framework with custom reward functions prioritizing safety over efficiency. Unity game engine-based simulation is used for direct collision modeling. A custom reward enabled reinforcement learning method, proximal policy optimization (PPO) model, yields substantial improvements over baseline results, reducing the number of serious collisions, number of vehicle-vehicle collisions, and total distance travelled by over 3 times the baseline values. The model also improves fuel efficiency by 39% and reduces carbon emissions by 88%. Results establish feasibility for city-wide 3D traffic simulation applications incorporating the vision-zero safety principles of the Department of Transportation, including physics-informed, adaptable, realistic collision modeling, as well as appropriate reward modeling for real-world traffic signal light control towards reducing collisions, optimizing traffic flow and reducing greenhouse emissions. | [
{
"version": "v1",
"created": "Fri, 23 May 2025 00:43:14 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Nuthakki",
"Mira",
""
]
] |
2506.03162 | Damith Chamalke Senadeera Mr | Damith Chamalke Senadeera, Xiaoyun Yang, Dimitrios Kollias, Gregory Slabaugh | Dual Branch VideoMamba with Gated Class Token Fusion for Violence Detection | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | The rapid proliferation of surveillance cameras has increased the demand for automated violence detection. While CNNs and Transformers have shown success in extracting spatio-temporal features, they struggle with long-term dependencies and computational efficiency. We propose Dual Branch VideoMamba with Gated Class Token Fusion (GCTF), an efficient architecture combining a dual-branch design and a state-space model (SSM) backbone where one branch captures spatial features, while the other focuses on temporal dynamics, with continuous fusion via a gating mechanism. We also present a new benchmark by merging RWF-2000, RLVS, and VioPeru datasets in video violence detection, ensuring strict separation between training and testing sets. Our model achieves state-of-the-art performance on this benchmark offering an optimal balance between accuracy and computational efficiency, demonstrating the promise of SSMs for scalable, real-time surveillance violence detection. | [
{
"version": "v1",
"created": "Fri, 23 May 2025 10:41:45 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Senadeera",
"Damith Chamalke",
""
],
[
"Yang",
"Xiaoyun",
""
],
[
"Kollias",
"Dimitrios",
""
],
[
"Slabaugh",
"Gregory",
""
]
] |
2506.03163 | Oluwaseyi Giwa | Oluwaseyi Giwa | Causal Discovery in Dynamic Fading Wireless Networks | 5 pages, 3 figures | null | null | null | cs.LG eess.SP stat.ME | http://creativecommons.org/licenses/by/4.0/ | Dynamic causal discovery in wireless networks is essential due to evolving interference, fading, and mobility, which complicate traditional static causal models. This paper addresses causal inference challenges in dynamic fading wireless environments by proposing a sequential regression-based algorithm with a novel application of the NOTEARS acyclicity constraint, enabling efficient online updates. We derive theoretical lower and upper bounds on the detection delay required to identify structural changes, explicitly quantifying their dependence on network size, noise variance, and fading severity. Monte Carlo simulations validate these theoretical results, demonstrating linear increases in detection delay with network size, quadratic growth with noise variance, and inverse-square dependence on the magnitude of structural changes. Our findings provide rigorous theoretical insights and practical guidelines for designing robust online causal inference mechanisms to maintain network reliability under nonstationary wireless conditions. | [
{
"version": "v1",
"created": "Sat, 24 May 2025 13:59:48 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Giwa",
"Oluwaseyi",
""
]
] |
2506.03164 | Vignav Ramesh | Vignav Ramesh, Morteza Mardani | Test-Time Scaling of Diffusion Models via Noise Trajectory Search | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | The iterative and stochastic nature of diffusion models enables test-time scaling, whereby spending additional compute during denoising generates higher-fidelity samples. Increasing the number of denoising steps is the primary scaling axis, but this yields quickly diminishing returns. Instead optimizing the noise trajectory--the sequence of injected noise vectors--is promising, as the specific noise realizations critically affect sample quality; but this is challenging due to a high-dimensional search space, complex noise-outcome interactions, and costly trajectory evaluations. We address this by first casting diffusion as a Markov Decision Process (MDP) with a terminal reward, showing tree-search methods such as Monte Carlo tree search (MCTS) to be meaningful but impractical. To balance performance and efficiency, we then resort to a relaxation of MDP, where we view denoising as a sequence of independent contextual bandits. This allows us to introduce an $\epsilon$-greedy search algorithm that globally explores at extreme timesteps and locally exploits during the intermediate steps where de-mixing occurs. Experiments on EDM and Stable Diffusion reveal state-of-the-art scores for class-conditioned/text-to-image generation, exceeding baselines by up to $164\%$ and matching/exceeding MCTS performance. To our knowledge, this is the first practical method for test-time noise trajectory optimization of arbitrary (non-differentiable) rewards. | [
{
"version": "v1",
"created": "Sat, 24 May 2025 19:13:29 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Ramesh",
"Vignav",
""
],
[
"Mardani",
"Morteza",
""
]
] |
2506.03165 | Brady Lund | Brady D. Lund and Ting Wang | What Does Information Science Offer for Data Science Research?: A Review of Data and Information Ethics Literature | null | Journal of Data and Information Science (2022) | 10.2478/jdis-2022-0018 | null | cs.DL cs.CY | http://creativecommons.org/licenses/by/4.0/ | This paper reviews literature pertaining to the development of data science as a discipline, current issues with data bias and ethics, and the role that the discipline of information science may play in addressing these concerns. Information science research and researchers have much to offer for data science, owing to their background as transdisciplinary scholars who apply human-centered and social-behavioral perspectives to issues within natural science disciplines. Information science researchers have already contributed to a humanistic approach to data ethics within the literature and an emphasis on data science within information schools all but ensures that this literature will continue to grow in coming decades. This review article serves as a reference for the history, current progress, and potential future directions of data ethics research within the corpus of information science literature. | [
{
"version": "v1",
"created": "Mon, 26 May 2025 14:07:42 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Lund",
"Brady D.",
""
],
[
"Wang",
"Ting",
""
]
] |
2506.03166 | Dimitrios Kafetzis | Dimitrios Kafetzis, Nikos Fotiou, Savvas Argyropoulos, Jad Nasreddine and Iordanis Koutsopoulos | Video Quality Monitoring for Remote Autonomous Vehicle Control | 7 pages | null | null | null | cs.NI cs.SY eess.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The delivery of high-quality, low-latency video streams is critical for remote autonomous vehicle control, where operators must intervene in real time. However, reliable video delivery over Fourth/Fifth-Generation (4G/5G) mobile networks is challenging due to signal variability, mobility-induced handovers, and transient congestion. In this paper, we present a comprehensive blueprint for an integrated video quality monitoring system, tailored to remote autonomous vehicle operation. Our proposed system includes subsystems for data collection onboard the vehicle, video capture and compression, data transmission to edge servers, real-time streaming data management, Artificial Intelligence (AI) model deployment and inference execution, and proactive decision-making based on predicted video quality. The AI models are trained on a hybrid dataset that combines field-trial measurements with synthetic stress segments and covers Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and encoder-only Transformer architectures. As a proof of concept, we benchmark 20 variants from these model classes together with feed-forward Deep Neural Network (DNN) and linear-regression baselines, reporting accuracy and inference latency. Finally, we study the trade-offs between onboard and edge-based inference. We further discuss the use of explainable AI techniques to enhance transparency and accountability during critical remote-control interventions. Our proactive approach to network adaptation and Quality of Experience (QoE) monitoring aims to enhance remote vehicle operation over next-generation wireless networks. | [
{
"version": "v1",
"created": "Tue, 27 May 2025 21:37:42 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Kafetzis",
"Dimitrios",
""
],
[
"Fotiou",
"Nikos",
""
],
[
"Argyropoulos",
"Savvas",
""
],
[
"Nasreddine",
"Jad",
""
],
[
"Koutsopoulos",
"Iordanis",
""
]
] |
2506.03167 | Long Le Tan | Long Tan Le, Senura Hansaja Wanasekara, Zerun Niu, Yansong Shi, Nguyen H. Tran, Phuong Vo, Walid Saad, Dusit Niyato, Zhu Han, Choong Seon Hong, H. Vincent Poor | Distributionally Robust Wireless Semantic Communication with Large AI Models | Under Review | null | null | null | cs.NI cs.ET cs.IT cs.LG math.IT | http://creativecommons.org/licenses/by-sa/4.0/ | 6G wireless systems are expected to support massive volumes of data with ultra-low latency. However, conventional bit-level transmission strategies cannot support the efficiency and adaptability required by modern, data-intensive applications. The concept of semantic communication (SemCom) addresses this limitation by focusing on transmitting task-relevant semantic information instead of raw data. While recent efforts incorporating deep learning and large-scale AI models have improved SemCom's performance, existing systems remain vulnerable to both semantic-level and transmission-level noise because they often rely on domain-specific architectures that hinder generalizability. In this paper, a novel and generalized semantic communication framework called WaSeCom is proposed to systematically address uncertainty and enhance robustness. In particular, Wasserstein distributionally robust optimization is employed to provide resilience against semantic misinterpretation and channel perturbations. A rigorous theoretical analysis is performed to establish the robust generalization guarantees of the proposed framework. Experimental results on image and text transmission demonstrate that WaSeCom achieves improved robustness under noise and adversarial perturbations. These results highlight its effectiveness in preserving semantic fidelity across varying wireless conditions. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 04:03:57 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Le",
"Long Tan",
""
],
[
"Wanasekara",
"Senura Hansaja",
""
],
[
"Niu",
"Zerun",
""
],
[
"Shi",
"Yansong",
""
],
[
"Tran",
"Nguyen H.",
""
],
[
"Vo",
"Phuong",
""
],
[
"Saad",
"Walid",
""
],
[
"Niyato",
"Dusit",
""
],
[
"Han",
"Zhu",
""
],
[
"Hong",
"Choong Seon",
""
],
[
"Poor",
"H. Vincent",
""
]
] |
2506.03168 | Zhishu Shen | Dawen Jiang, Zhishu Shen, Qiushi Zheng, Tiehua Zhang, Wei Xiang, Jiong Jin | Farm-LightSeek: An Edge-centric Multimodal Agricultural IoT Data Analytics Framework with Lightweight LLMs | Accepted by IEEE Internet of Things Magazine | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Amid the challenges posed by global population growth and climate change, traditional agricultural Internet of Things (IoT) systems is currently undergoing a significant digital transformation to facilitate efficient big data processing. While smart agriculture utilizes artificial intelligence (AI) technologies to enable precise control, it still encounters significant challenges, including excessive reliance on agricultural expert knowledge, difficulties in fusing multimodal data, poor adaptability to dynamic environments, and bottlenecks in real-time decision-making at the edge. Large language models (LLMs), with their exceptional capabilities in knowledge acquisition and semantic understanding, provide a promising solution to address these challenges. To this end, we propose Farm-LightSeek, an edge-centric multimodal agricultural IoT data analytics framework that integrates LLMs with edge computing. This framework collects real-time farmland multi-source data (images, weather, geographic information) via sensors, performs cross-modal reasoning and disease detection at edge nodes, conducts low-latency management decisions, and enables cloud collaboration for model updates. The main innovations of Farm-LightSeek include: (1) an agricultural "perception-decision-action" closed-loop architecture; (2) cross-modal adaptive monitoring; and (3)a lightweight LLM deployment strategy balancing performance and efficiency. Experiments conducted on two real-world datasets demonstrate that Farm-LightSeek consistently achieves reliable performance in mission-critical tasks, even under the limitations of edge computing resources. This work advances intelligent real-time agricultural solutions and highlights the potential for deeper integration of agricultural IoT with LLMs. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 14:09:36 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Jiang",
"Dawen",
""
],
[
"Shen",
"Zhishu",
""
],
[
"Zheng",
"Qiushi",
""
],
[
"Zhang",
"Tiehua",
""
],
[
"Xiang",
"Wei",
""
],
[
"Jin",
"Jiong",
""
]
] |
2506.03169 | Arindam Chaudhuri Arc | Arindam Chaudhuri | Improvement of human health lifespan with hybrid group pose estimation methods | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/publicdomain/zero/1.0/ | Human beings rely heavily on estimation of poses in order to access their body movements. Human pose estimation methods take advantage of computer vision advances in order to track human body movements in real life applications. This comes from videos which are recorded through available devices. These para-digms provide potential to make human movement measurement more accessible to users. The consumers of pose estimation movements believe that human poses content tend to supplement available videos. This has increased pose estimation software usage to estimate human poses. In order to address this problem, we develop hybrid-ensemble-based group pose estimation method to improve human health. This proposed hybrid-ensemble-based group pose estimation method aims to detect multi-person poses using modified group pose estimation and modified real time pose estimation. This ensemble allows fusion of performance of stated methods in real time. The input poses from images are fed into individual meth-ods. The pose transformation method helps to identify relevant features for en-semble to perform training effectively. After this, customized pre-trained hybrid ensemble is trained on public benchmarked datasets which is being evaluated through test datasets. The effectiveness and viability of proposed method is estab-lished based on comparative analysis of group pose estimation methods and ex-periments conducted on benchmarked datasets. It provides best optimized results in real-time pose estimation. It makes pose estimation method more robust to oc-clusion and improves dense regression accuracy. These results have affirmed po-tential application of this method in several real-time situations with improvement in human health life span | [
{
"version": "v1",
"created": "Wed, 28 May 2025 16:43:28 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chaudhuri",
"Arindam",
""
]
] |
2506.03170 | Murthy L | Murthy L, Subarna Tripathi | PALADIN : Robust Neural Fingerprinting for Text-to-Image Diffusion Models | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | The risk of misusing text-to-image generative models for malicious uses, especially due to the open-source development of such models, has become a serious concern. As a risk mitigation strategy, attributing generative models with neural fingerprinting is emerging as a popular technique. There has been a plethora of recent work that aim for addressing neural fingerprinting. A trade-off between the attribution accuracy and generation quality of such models has been studied extensively. None of the existing methods yet achieved $100\%$ attribution accuracy. However, any model with less than \emph{perfect} accuracy is practically non-deployable. In this work, we propose an accurate method to incorporate neural fingerprinting for text-to-image diffusion models leveraging the concepts of cyclic error correcting codes from the literature of coding theory. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 18:52:40 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"L",
"Murthy",
""
],
[
"Tripathi",
"Subarna",
""
]
] |
2506.03171 | Ghulam Mujtaba | Ghulam Mujtaba, Eun-Seok Ryu | EdgeVidSum: Real-Time Personalized Video Summarization at the Edge | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | EdgeVidSum is a lightweight method that generates personalized, fast-forward summaries of long-form videos directly on edge devices. The proposed approach enables real-time video summarization while safeguarding user privacy through local data processing using innovative thumbnail-based techniques and efficient neural architectures. Unlike conventional methods that process entire videos frame by frame, the proposed method uses thumbnail containers to significantly reduce computational complexity without sacrificing semantic relevance. The framework employs a hierarchical analysis approach, where a lightweight 2D CNN model identifies user-preferred content from thumbnails and generates timestamps to create fast-forward summaries. Our interactive demo highlights the system's ability to create tailored video summaries for long-form videos, such as movies, sports events, and TV shows, based on individual user preferences. The entire computation occurs seamlessly on resource-constrained devices like Jetson Nano, demonstrating how EdgeVidSum addresses the critical challenges of computational efficiency, personalization, and privacy in modern video consumption environments. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 18:59:41 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Mujtaba",
"Ghulam",
""
],
[
"Ryu",
"Eun-Seok",
""
]
] |
2506.03172 | Jingyi Zhao | Jingyi Zhao, Claudia Archetti, Tuan Anh Pham, Thibaut Vidal | Large Neighborhood and Hybrid Genetic Search for Inventory Routing Problems | 24 pages | null | null | null | cs.NE stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The inventory routing problem (IRP) focuses on jointly optimizing inventory and distribution operations from a supplier to retailers over multiple days. Compared to other problems from the vehicle routing family, the interrelations between inventory and routing decisions render IRP optimization more challenging and call for advanced solution techniques. A few studies have focused on developing large neighborhood search approaches for this class of problems, but this remains a research area with vast possibilities due to the challenges related to the integration of inventory and routing decisions. In this study, we advance this research area by developing a new large neighborhood search operator tailored for the IRP. Specifically, the operator optimally removes and reinserts all visits to a specific retailer while minimizing routing and inventory costs. We propose an efficient tailored dynamic programming algorithm that exploits preprocessing and acceleration strategies. The operator is used to build an effective local search routine, and included in a state-of-the-art routing algorithm, i.e., Hybrid Genetic Search (HGS). Through extensive computational experiments, we demonstrate that the resulting heuristic algorithm leads to solutions of unmatched quality up to this date, especially on large-scale benchmark instances. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 21:18:08 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zhao",
"Jingyi",
""
],
[
"Archetti",
"Claudia",
""
],
[
"Pham",
"Tuan Anh",
""
],
[
"Vidal",
"Thibaut",
""
]
] |
2506.03174 | Tsuyoshi Okita | Koki Matsuishi and Kosuke Ukita and Tsuyoshi Okita | Multimodal Foundation Model for Cross-Modal Retrieval and Activity Recognition Tasks | 25 pages, 8 figures | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, the widespread adoption of wearable devices has highlighted the growing importance of behavior analysis using IMU. While applications span diverse fields such as healthcare and robotics, recent studies have increasingly focused on multimodal analysis, in addition to unimodal analysis. Several studies have proposed multimodal foundation models that incorporate first-person video and text data; however, these models still fall short in providing a detailed analysis of full-body human activity. To address this limitation, we propose Activity Understanding and Representations Alignment - Multimodal Foundation Model (AURA-MFM), a foundational model integrating four modalities: third-person video, motion capture, IMU, and text. By incorporating third-person video and motion capture data, the model enables a detailed and multidimensional understanding of human activity, which first-person perspectives alone fail to capture. Additionally, a Transformer-based IMU encoder is employed to enhance the model's overall performance. Experimental evaluations on retrieval and activity recognition tasks demonstrate that our model surpasses existing methods. Notably, in the zero-shot classification for action recognition, our method achieved significantly higher performance, with an F1-score of 0.6226 and an accuracy of 0.7320, whereas the existing method recorded an F1-score of 0.0747 and an accuracy of 0.1961. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 01:47:43 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Matsuishi",
"Koki",
""
],
[
"Ukita",
"Kosuke",
""
],
[
"Okita",
"Tsuyoshi",
""
]
] |
2506.03175 | Xiao Youshen | Youshen Xiao, Yiling Shi, Ruixi Sun, Hongjiang Wei, Fei Gao, Yuyao Zhang | Super-temporal-resolution Photoacoustic Imaging with Dynamic Reconstruction through Implicit Neural Representation in Sparse-view | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Dynamic Photoacoustic Computed Tomography (PACT) is an important imaging technique for monitoring physiological processes, capable of providing high-contrast images of optical absorption at much greater depths than traditional optical imaging methods. However, practical instrumentation and geometric constraints limit the number of acoustic sensors available around the imaging target, leading to sparsity in sensor data. Traditional photoacoustic (PA) image reconstruction methods, when directly applied to sparse PA data, produce severe artifacts. Additionally, these traditional methods do not consider the inter-frame relationships in dynamic imaging. Temporal resolution is crucial for dynamic photoacoustic imaging, which is fundamentally limited by the low repetition rate (e.g., 20 Hz) and high cost of high-power laser technology. Recently, Implicit Neural Representation (INR) has emerged as a powerful deep learning tool for solving inverse problems with sparse data, by characterizing signal properties as continuous functions of their coordinates in an unsupervised manner. In this work, we propose an INR-based method to improve dynamic photoacoustic image reconstruction from sparse-views and enhance temporal resolution, using only spatiotemporal coordinates as input. Specifically, the proposed INR represents dynamic photoacoustic images as implicit functions and encodes them into a neural network. The weights of the network are learned solely from the acquired sparse sensor data, without the need for external training datasets or prior images. Benefiting from the strong implicit continuity regularization provided by INR, as well as explicit regularization for low-rank and sparsity, our proposed method outperforms traditional reconstruction methods under two different sparsity conditions, effectively suppressing artifacts and ensuring image quality. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 06:36:44 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Xiao",
"Youshen",
""
],
[
"Shi",
"Yiling",
""
],
[
"Sun",
"Ruixi",
""
],
[
"Wei",
"Hongjiang",
""
],
[
"Gao",
"Fei",
""
],
[
"Zhang",
"Yuyao",
""
]
] |
2506.03176 | Yongqi Han | Bin Wang, Yongqi Han, Minbo Ma, Tianrui Li, Junbo Zhang, Feng Hong, Yanwei Yu | Non-collective Calibrating Strategy for Time Series Forecasting | Accepted by IJCAI 2025 | null | null | null | cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Deep learning-based approaches have demonstrated significant advancements in time series forecasting. Despite these ongoing developments, the complex dynamics of time series make it challenging to establish the rule of thumb for designing the golden model architecture. In this study, we argue that refining existing advanced models through a universal calibrating strategy can deliver substantial benefits with minimal resource costs, as opposed to elaborating and training a new model from scratch. We first identify a multi-target learning conflict in the calibrating process, which arises when optimizing variables across time steps, leading to the underutilization of the model's learning capabilities. To address this issue, we propose an innovative calibrating strategy called Socket+Plug (SoP). This approach retains an exclusive optimizer and early-stopping monitor for each predicted target within each Plug while keeping the fully trained Socket backbone frozen. The model-agnostic nature of SoP allows it to directly calibrate the performance of any trained deep forecasting models, regardless of their specific architectures. Extensive experiments on various time series benchmarks and a spatio-temporal meteorological ERA5 dataset demonstrate the effectiveness of SoP, achieving up to a 22% improvement even when employing a simple MLP as the Plug (highlighted in Figure 1) | [
{
"version": "v1",
"created": "Thu, 29 May 2025 10:14:26 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wang",
"Bin",
""
],
[
"Han",
"Yongqi",
""
],
[
"Ma",
"Minbo",
""
],
[
"Li",
"Tianrui",
""
],
[
"Zhang",
"Junbo",
""
],
[
"Hong",
"Feng",
""
],
[
"Yu",
"Yanwei",
""
]
] |
2506.03178 | Ashad Kabir | Md. Zihad Bin Jahangir, Muhammad Ashad Kabir, Sumaiya Akter, Israt Jahan, Minh Chau | LLaMA-XR: A Novel Framework for Radiology Report Generation using LLaMA and QLoRA Fine Tuning | 25 pages | null | null | null | eess.IV cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Automated radiology report generation holds significant potential to reduce radiologists' workload and enhance diagnostic accuracy. However, generating precise and clinically meaningful reports from chest radiographs remains challenging due to the complexity of medical language and the need for contextual understanding. Existing models often struggle with maintaining both accuracy and contextual relevance. In this paper, we present LLaMA-XR, a novel framework that integrates LLaMA 3.1 with DenseNet-121-based image embeddings and Quantized Low-Rank Adaptation (QLoRA) fine-tuning. LLaMA-XR achieves improved coherence and clinical accuracy while maintaining computational efficiency. This efficiency is driven by an optimization strategy that enhances parameter utilization and reduces memory overhead, enabling faster report generation with lower computational resource demands. Extensive experiments conducted on the IU X-ray benchmark dataset demonstrate that LLaMA-XR outperforms a range of state-of-the-art methods. Our model achieves a ROUGE-L score of 0.433 and a METEOR score of 0.336, establishing new performance benchmarks in the domain. These results underscore LLaMA-XR's potential as an effective and efficient AI system for automated radiology reporting, offering enhanced clinical utility and reliability. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 12:21:18 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Jahangir",
"Md. Zihad Bin",
""
],
[
"Kabir",
"Muhammad Ashad",
""
],
[
"Akter",
"Sumaiya",
""
],
[
"Jahan",
"Israt",
""
],
[
"Chau",
"Minh",
""
]
] |
2506.03179 | Qi Li | Qi Li, Runpeng Yu, Xinchao Wang | Vid-SME: Membership Inference Attacks against Large Video Understanding Models | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Multimodal large language models (MLLMs) demonstrate remarkable capabilities in handling complex multimodal tasks and are increasingly adopted in video understanding applications. However, their rapid advancement raises serious data privacy concerns, particularly given the potential inclusion of sensitive video content, such as personal recordings and surveillance footage, in their training datasets. Determining improperly used videos during training remains a critical and unresolved challenge. Despite considerable progress on membership inference attacks (MIAs) for text and image data in MLLMs, existing methods fail to generalize effectively to the video domain. These methods suffer from poor scalability as more frames are sampled and generally achieve negligible true positive rates at low false positive rates (TPR@Low FPR), mainly due to their failure to capture the inherent temporal variations of video frames and to account for model behavior differences as the number of frames varies. To address these challenges, we introduce Vid-SME, the first membership inference method tailored for video data used in video understanding LLMs (VULLMs). Vid-SME leverages the confidence of model output and integrates adaptive parameterization to compute Sharma-Mittal entropy (SME) for video inputs. By leveraging the SME difference between natural and temporally-reversed video frames, Vid-SME derives robust membership scores to determine whether a given video is part of the model's training set. Experiments on various self-trained and open-sourced VULLMs demonstrate the strong effectiveness of Vid-SME. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 13:17:25 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Li",
"Qi",
""
],
[
"Yu",
"Runpeng",
""
],
[
"Wang",
"Xinchao",
""
]
] |
2506.03180 | Jan Ignatowicz | Jan Ignatowicz, Krzysztof Kutt, Grzegorz J. Nalepa | Knowledge Graphs for Digitized Manuscripts in Jagiellonian Digital Library Application | null | null | null | null | cs.DL cs.CV | http://creativecommons.org/licenses/by/4.0/ | Digitizing cultural heritage collections has become crucial for preservation of historical artifacts and enhancing their availability to the wider public. Galleries, libraries, archives and museums (GLAM institutions) are actively digitizing their holdings and creates extensive digital collections. Those collections are often enriched with metadata describing items but not exactly their contents. The Jagiellonian Digital Library, standing as a good example of such an effort, offers datasets accessible through protocols like OAI-PMH. Despite these improvements, metadata completeness and standardization continue to pose substantial obstacles, limiting the searchability and potential connections between collections. To deal with these challenges, we explore an integrated methodology of computer vision (CV), artificial intelligence (AI), and semantic web technologies to enrich metadata and construct knowledge graphs for digitized manuscripts and incunabula. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 14:49:24 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Ignatowicz",
"Jan",
""
],
[
"Kutt",
"Krzysztof",
""
],
[
"Nalepa",
"Grzegorz J.",
""
]
] |
2506.03181 | Wangting Zhou | Wangting Zhou, Jiangshan He, Tong Cai, Lin Wang, Zhen Yuan, Xunbin Wei, Xueli Chen | Dc-EEMF: Pushing depth-of-field limit of photoacoustic microscopy via decision-level constrained learning | null | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Photoacoustic microscopy holds the potential to measure biomarkers' structural and functional status without labels, which significantly aids in comprehending pathophysiological conditions in biomedical research. However, conventional optical-resolution photoacoustic microscopy (OR-PAM) is hindered by a limited depth-of-field (DoF) due to the narrow depth range focused on a Gaussian beam. Consequently, it fails to resolve sufficient details in the depth direction. Herein, we propose a decision-level constrained end-to-end multi-focus image fusion (Dc-EEMF) to push DoF limit of PAM. The DC-EEMF method is a lightweight siamese network that incorporates an artifact-resistant channel-wise spatial frequency as its feature fusion rule. The meticulously crafted U-Net-based perceptual loss function for decision-level focus properties in end-to-end fusion seamlessly integrates the complementary advantages of spatial domain and transform domain methods within Dc-EEMF. This approach can be trained end-to-end without necessitating post-processing procedures. Experimental results and numerical analyses collectively demonstrate our method's robust performance, achieving an impressive fusion result for PAM images without a substantial sacrifice in lateral resolution. The utilization of Dc-EEMF-powered PAM has the potential to serve as a practical tool in preclinical and clinical studies requiring extended DoF for various applications. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 15:17:52 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zhou",
"Wangting",
""
],
[
"He",
"Jiangshan",
""
],
[
"Cai",
"Tong",
""
],
[
"Wang",
"Lin",
""
],
[
"Yuan",
"Zhen",
""
],
[
"Wei",
"Xunbin",
""
],
[
"Chen",
"Xueli",
""
]
] |
2506.03182 | Chinmay Hegde | Shivani Chiranjeevi and Hossein Zaremehrjerdi and Zi K. Deng and Talukder Z. Jubery and Ari Grele and Arti Singh and Asheesh K Singh and Soumik Sarkar and Nirav Merchant and Harold F. Greeney and Baskar Ganapathysubramanian and Chinmay Hegde | TerraIncognita: A Dynamic Benchmark for Species Discovery Using Frontier Models | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | The rapid global loss of biodiversity, particularly among insects, represents an urgent ecological crisis. Current methods for insect species discovery are manual, slow, and severely constrained by taxonomic expertise, hindering timely conservation actions. We introduce TerraIncognita, a dynamic benchmark designed to evaluate state-of-the-art multimodal models for the challenging problem of identifying unknown, potentially undescribed insect species from image data. Our benchmark dataset combines a mix of expertly annotated images of insect species likely known to frontier AI models, and images of rare and poorly known species, for which few/no publicly available images exist. These images were collected from underexplored biodiversity hotspots, realistically mimicking open-world discovery scenarios faced by ecologists. The benchmark assesses models' proficiency in hierarchical taxonomic classification, their capability to detect and abstain from out-of-distribution (OOD) samples representing novel species, and their ability to generate explanations aligned with expert taxonomic knowledge. Notably, top-performing models achieve over 90\% F1 at the Order level on known species, but drop below 2\% at the Species level, highlighting the sharp difficulty gradient from coarse to fine taxonomic prediction (Order $\rightarrow$ Family $\rightarrow$ Genus $\rightarrow$ Species). TerraIncognita will be updated regularly, and by committing to quarterly dataset expansions (of both known and novel species), will provide an evolving platform for longitudinal benchmarking of frontier AI methods. All TerraIncognita data, results, and future updates are available \href{https://baskargroup.github.io/TerraIncognita/}{here}. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 15:20:15 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Chiranjeevi",
"Shivani",
""
],
[
"Zaremehrjerdi",
"Hossein",
""
],
[
"Deng",
"Zi K.",
""
],
[
"Jubery",
"Talukder Z.",
""
],
[
"Grele",
"Ari",
""
],
[
"Singh",
"Arti",
""
],
[
"Singh",
"Asheesh K",
""
],
[
"Sarkar",
"Soumik",
""
],
[
"Merchant",
"Nirav",
""
],
[
"Greeney",
"Harold F.",
""
],
[
"Ganapathysubramanian",
"Baskar",
""
],
[
"Hegde",
"Chinmay",
""
]
] |
2506.03183 | Ya\c{s}ar Utku Al\c{c}alar | Ya\c{s}ar Utku Al\c{c}alar, Yu Cao, Mehmet Ak\c{c}akaya | Edge Computing for Physics-Driven AI in Computational MRI: A Feasibility Study | IEEE International Conference on Future Internet of Things and Cloud (FiCloud), 2025 | null | null | null | eess.IV cs.AI cs.AR cs.CV cs.LG physics.med-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Physics-driven artificial intelligence (PD-AI) reconstruction methods have emerged as the state-of-the-art for accelerating MRI scans, enabling higher spatial and temporal resolutions. However, the high resolution of these scans generates massive data volumes, leading to challenges in transmission, storage, and real-time processing. This is particularly pronounced in functional MRI, where hundreds of volumetric acquisitions further exacerbate these demands. Edge computing with FPGAs presents a promising solution for enabling PD-AI reconstruction near the MRI sensors, reducing data transfer and storage bottlenecks. However, this requires optimization of PD-AI models for hardware efficiency through quantization and bypassing traditional FFT-based approaches, which can be a limitation due to their computational demands. In this work, we propose a novel PD-AI computational MRI approach optimized for FPGA-based edge computing devices, leveraging 8-bit complex data quantization and eliminating redundant FFT/IFFT operations. Our results show that this strategy improves computational efficiency while maintaining reconstruction quality comparable to conventional PD-AI methods, and outperforms standard clinical methods. Our approach presents an opportunity for high-resolution MRI reconstruction on resource-constrained devices, highlighting its potential for real-world deployment. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 02:35:43 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Alçalar",
"Yaşar Utku",
""
],
[
"Cao",
"Yu",
""
],
[
"Akçakaya",
"Mehmet",
""
]
] |
2506.03184 | Md. Monirul Islam | Mahe Zabin, Ho-Jin Choi, Md. Monirul Islam, and Jia Uddin | Impact of Tuning Parameters in Deep Convolutional Neural Network Using a Crack Image Dataset | 8 pages, 2 figures, published at Proceedings of the 15th KIPS International Conference on Ubiquitous Information Technologies and Applications (CUTE 2021), Jeju, Repubilc of Korea | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | The performance of a classifier depends on the tuning of its parame ters. In this paper, we have experimented the impact of various tuning parameters on the performance of a deep convolutional neural network (DCNN). In the ex perimental evaluation, we have considered a DCNN classifier that consists of 2 convolutional layers (CL), 2 pooling layers (PL), 1 dropout, and a dense layer. To observe the impact of pooling, activation function, and optimizer tuning pa rameters, we utilized a crack image dataset having two classes: negative and pos itive. The experimental results demonstrate that with the maxpooling, the DCNN demonstrates its better performance for adam optimizer and tanh activation func tion. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 10:58:31 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Zabin",
"Mahe",
""
],
[
"Choi",
"Ho-Jin",
""
],
[
"Islam",
"Md. Monirul",
""
],
[
"Uddin",
"Jia",
""
]
] |
2506.03185 | Liangrui Pan | Liangrui Pan, Xingchen Li, Zhongyi Chen, Ling Chu, Shaoliang Peng | DLiPath: A Benchmark for the Comprehensive Assessment of Donor Liver Based on Histopathological Image Dataset | Submit to ACM MM2025 | null | null | null | eess.IV cs.AI cs.CV q-bio.QM | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Pathologists comprehensive evaluation of donor liver biopsies provides crucial information for accepting or discarding potential grafts. However, rapidly and accurately obtaining these assessments intraoperatively poses a significant challenge for pathologists. Features in donor liver biopsies, such as portal tract fibrosis, total steatosis, macrovesicular steatosis, and hepatocellular ballooning are correlated with transplant outcomes, yet quantifying these indicators suffers from substantial inter- and intra-observer variability. To address this, we introduce DLiPath, the first benchmark for comprehensive donor liver assessment based on a histopathology image dataset. We collected and publicly released 636 whole slide images from 304 donor liver patients at the Department of Pathology, the Third Xiangya Hospital, with expert annotations for key pathological features (including cholestasis, portal tract fibrosis, portal inflammation, total steatosis, macrovesicular steatosis, and hepatocellular ballooning). We selected nine state-of-the-art multiple-instance learning (MIL) models based on the DLiPath dataset as baselines for extensive comparative analysis. The experimental results demonstrate that several MIL models achieve high accuracy across donor liver assessment indicators on DLiPath, charting a clear course for future automated and intelligent donor liver assessment research. Data and code are available at https://github.com/panliangrui/ACM_MM_2025. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 12:13:00 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Pan",
"Liangrui",
""
],
[
"Li",
"Xingchen",
""
],
[
"Chen",
"Zhongyi",
""
],
[
"Chu",
"Ling",
""
],
[
"Peng",
"Shaoliang",
""
]
] |
2506.03186 | Sabah Abdulazeez | Duaa Kareem Qasim, Sabah Abdulazeez Jebur, Lafta Raheem Ali, Abdul Jalil M. Khalaf, Abir Jaafar Hussain | Lightweight Convolutional Neural Networks for Retinal Disease Classification | null | null | null | null | eess.IV cs.AI cs.CV cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Retinal diseases such as Diabetic Retinopathy (DR) and Macular Hole (MH) significantly impact vision and affect millions worldwide. Early detection is crucial, as DR, a complication of diabetes, damages retinal blood vessels, potentially leading to blindness, while MH disrupts central vision, affecting tasks like reading and facial recognition. This paper employed two lightweight and efficient Convolution Neural Network architectures, MobileNet and NASNetMobile, for the classification of Normal, DR, and MH retinal images. The models were trained on the RFMiD dataset, consisting of 3,200 fundus images, after undergoing preprocessing steps such as resizing, normalization, and augmentation. To address data scarcity, this study leveraged transfer learning and data augmentation techniques, enhancing model generalization and performance. The experimental results demonstrate that MobileNetV2 achieved the highest accuracy of 90.8%, outperforming NASNetMobile, which achieved 89.5% accuracy. These findings highlight the effectiveness of CNNs in retinal disease classification, providing a foundation for AI-assisted ophthalmic diagnosis and early intervention. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 12:36:45 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Qasim",
"Duaa Kareem",
""
],
[
"Jebur",
"Sabah Abdulazeez",
""
],
[
"Ali",
"Lafta Raheem",
""
],
[
"Khalaf",
"Abdul Jalil M.",
""
],
[
"Hussain",
"Abir Jaafar",
""
]
] |
2506.03187 | Malena M\'endez Isla | Malena Mendez Isla, Agustin Mauro and Diego Kozlowski | Comparing Retrieval Strategies to Capture Interdisciplinary Scientific Research: A Bibliometric Evaluation of the Integration of Neuroscience and Computer Science | null | null | null | null | cs.DL cs.IR | http://creativecommons.org/licenses/by/4.0/ | Interdisciplinary scientific research is increasingly important in knowledge production, funding policies, and academic discussions on scholarly communication. While many studies focus on interdisciplinary corpora defined a priori - usually through keyword-based searches within assumed interdisciplinary domains - few explore interdisciplinarity as an emergent intersection between two distinct fields. Thus, methodological proposals for building databases at the intersection of two fields of knowledge are scarce. The goal of this article is to develop and compare different strategies for defining an interdisciplinary corpus between two bodies of knowledge. As a case study, we focus on the intersection between neuroscience and computer science. To this end, we develop and compare four retrieval strategies, two of them based on keywords and two based on citation and reference patterns. Our results show that keyword-based strategies provide both better precision and recall. While we focus on comparing strategies for the study of the intersection between the fields of neuroscience and computer science, this proposed methodological reflection is applicable to a wide range of interdisciplinary domains. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 19:29:18 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Isla",
"Malena Mendez",
""
],
[
"Mauro",
"Agustin",
""
],
[
"Kozlowski",
"Diego",
""
]
] |
2506.03188 | Madhu Babu Sikha | Madhu Babu Sikha, Lalith Appari, Gurudatt Nanjanagudu Ganesh, Amay Bandodkar, Imon Banerjee | Multi-Analyte, Swab-based Automated Wound Monitor with AI | 4 pages conference paper | null | null | null | eess.IV cs.AI cs.CV cs.HC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Diabetic foot ulcers (DFUs), a class of chronic wounds, affect ~750,000 individuals every year in the US alone and identifying non-healing DFUs that develop to chronic wounds early can drastically reduce treatment costs and minimize risks of amputation. There is therefore a pressing need for diagnostic tools that can detect non-healing DFUs early. We develop a low cost, multi-analyte 3D printed assays seamlessly integrated on swabs that can identify non-healing DFUs and a Wound Sensor iOS App - an innovative mobile application developed for the controlled acquisition and automated analysis of wound sensor data. By comparing both the original base image (before exposure to the wound) and the wound-exposed image, we developed automated computer vision techniques to compare density changes between the two assay images, which allow us to automatically determine the severity of the wound. The iOS app ensures accurate data collection and presents actionable insights, despite challenges such as variations in camera configurations and ambient conditions. The proposed integrated sensor and iOS app will allow healthcare professionals to monitor wound conditions real-time, track healing progress, and assess critical parameters related to wound care. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 20:42:37 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Sikha",
"Madhu Babu",
""
],
[
"Appari",
"Lalith",
""
],
[
"Ganesh",
"Gurudatt Nanjanagudu",
""
],
[
"Bandodkar",
"Amay",
""
],
[
"Banerjee",
"Imon",
""
]
] |
2506.03189 | Ghada Sokar | Ghada Sokar, Gintare Karolina Dziugaite, Anurag Arnab, Ahmet Iscen, Pablo Samuel Castro, Cordelia Schmid | Continual Learning in Vision-Language Models via Aligned Model Merging | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Continual learning is conventionally tackled through sequential fine-tuning, a process that, while enabling adaptation, inherently favors plasticity over the stability needed to retain prior knowledge. While existing approaches attempt to mitigate catastrophic forgetting, a bias towards recent tasks persists as they build upon this sequential nature. In this work we present a new perspective based on model merging to maintain stability while still retaining plasticity. Rather than just sequentially updating the model weights, we propose merging newly trained task parameters with previously learned ones, promoting a better balance. To maximize the effectiveness of the merging process, we propose a simple mechanism that promotes learning aligned weights with previous ones, thereby avoiding interference when merging. We evaluate this approach on large Vision-Language Models (VLMs), and demonstrate its effectiveness in reducing forgetting, increasing robustness to various task orders and similarities, and improving generalization. | [
{
"version": "v1",
"created": "Fri, 30 May 2025 20:52:21 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Sokar",
"Ghada",
""
],
[
"Dziugaite",
"Gintare Karolina",
""
],
[
"Arnab",
"Anurag",
""
],
[
"Iscen",
"Ahmet",
""
],
[
"Castro",
"Pablo Samuel",
""
],
[
"Schmid",
"Cordelia",
""
]
] |
2506.03190 | Jamie Yi | Jiaming Yi, Ruirui Pan, Jishen Yang and Xiulong Yang | MINT: Memory-Infused Prompt Tuning at Test-time for CLIP | 14 pages, 3 figures | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Improving the generalization ability of Vision-Language Pre-trained Models (VLMs) under test-time data distribution shifts remains a critical challenge. The existing Test-Time Adaptation (TTA) methods fall short in fully leveraging the model's internal knowledge, particularly in dynamically adapting to complex and hierarchical visual semantic information. In this paper, we propose Memory-Infused Prompt Tuning (MINT), a novel framework to address this issue. Inspired by human associative memory theory, MINT introduces a Memory Prompt Bank (MPB), which stores learnable key-value prompt pairs that work as a memory of previously seen samples. During the test time, relevant prompt pairs in the MPB are retrieved by the hierarchical visual features of test images to dynamically assemble Associative Prompts. The associative prompts are then injected into the image encoder for fine-grained, customized visual contextual guidance. MINT also utilizes learnable text prompts. MINT thus enables rapid, precise VLM adaptation at test time by leveraging this MPB-acquired memory, without source data or retraining. The code is available at https://github.com/Jamieyi2004/MINT. | [
{
"version": "v1",
"created": "Sat, 31 May 2025 07:31:20 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yi",
"Jiaming",
""
],
[
"Pan",
"Ruirui",
""
],
[
"Yang",
"Jishen",
""
],
[
"Yang",
"Xiulong",
""
]
] |
2506.03191 | Muhammad Islam | Muhammad Islam, Tao Huang, Euijoon Ahn, and Usman Naseem | Multimodal Generative AI with Autoregressive LLMs for Human Motion Understanding and Generation: A Way Forward | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents an in-depth survey on the use of multimodal Generative Artificial Intelligence (GenAI) and autoregressive Large Language Models (LLMs) for human motion understanding and generation, offering insights into emerging methods, architectures, and their potential to advance realistic and versatile motion synthesis. Focusing exclusively on text and motion modalities, this research investigates how textual descriptions can guide the generation of complex, human-like motion sequences. The paper explores various generative approaches, including autoregressive models, diffusion models, Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and transformer-based models, by analyzing their strengths and limitations in terms of motion quality, computational efficiency, and adaptability. It highlights recent advances in text-conditioned motion generation, where textual inputs are used to control and refine motion outputs with greater precision. The integration of LLMs further enhances these models by enabling semantic alignment between instructions and motion, improving coherence and contextual relevance. This systematic survey underscores the transformative potential of text-to-motion GenAI and LLM architectures in applications such as healthcare, humanoids, gaming, animation, and assistive technologies, while addressing ongoing challenges in generating efficient and realistic human motion. | [
{
"version": "v1",
"created": "Sat, 31 May 2025 11:02:24 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Islam",
"Muhammad",
""
],
[
"Huang",
"Tao",
""
],
[
"Ahn",
"Euijoon",
""
],
[
"Naseem",
"Usman",
""
]
] |
2506.03192 | Basudha Pal | Basudha Pal, Rama Chellappa, Muhammad Umair | Encoding of Demographic and Anatomical Information in Chest X-Ray-based Severe Left Ventricular Hypertrophy Classifiers | null | null | null | null | eess.IV cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | While echocardiography and MRI are clinical standards for evaluating cardiac structure, their use is limited by cost and accessibility.We introduce a direct classification framework that predicts severe left ventricular hypertrophy from chest X-rays, without relying on anatomical measurements or demographic inputs. Our approach achieves high AUROC and AUPRC, and employs Mutual Information Neural Estimation to quantify feature expressivity. This reveals clinically meaningful attribute encoding and supports transparent model interpretation. | [
{
"version": "v1",
"created": "Sat, 31 May 2025 13:30:04 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Pal",
"Basudha",
""
],
[
"Chellappa",
"Rama",
""
],
[
"Umair",
"Muhammad",
""
]
] |
2506.03193 | Ekram Alam | Ekram Alam, Abu Sufian, Paramartha Dutta, Marco Leo | Human Fall Detection using Transfer Learning-based 3D CNN | null | null | 10.1007/978-3-031-81935-3_9 | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Unintentional or accidental falls are one of the significant health issues in senior persons. The population of senior persons is increasing steadily. So, there is a need for an automated fall detection monitoring system. This paper introduces a vision-based fall detection system using a pre-trained 3D CNN. Unlike 2D CNN, 3D CNN extracts not only spatial but also temporal features. The proposed model leverages the original learned weights of a 3D CNN model pre-trained on the Sports1M dataset to extract the spatio-temporal features. Only the SVM classifier was trained, which saves the time required to train the 3D CNN. Stratified shuffle five split cross-validation has been used to split the dataset into training and testing data. Extracted features from the proposed 3D CNN model were fed to an SVM classifier to classify the activity as fall or ADL. Two datasets, GMDCSA and CAUCAFall, were utilized to conduct the experiment. The source code for this work can be accessed via the following link: https://github.com/ekramalam/HFD_3DCNN. | [
{
"version": "v1",
"created": "Sat, 31 May 2025 16:58:12 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Alam",
"Ekram",
""
],
[
"Sufian",
"Abu",
""
],
[
"Dutta",
"Paramartha",
""
],
[
"Leo",
"Marco",
""
]
] |
2506.03194 | Sahiti Yerramilli | Rynaa Grover, Jayant Sravan Tamarapalli, Sahiti Yerramilli, Nilay Pande | HueManity: Probing Fine-Grained Visual Perception in MLLMs | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Multimodal Large Language Models (MLLMs) excel at high-level visual reasoning, but their performance on nuanced perceptual tasks remains surprisingly limited. We present HueManity, a benchmark designed to assess visual perception in MLLMs. The dataset comprises 83,850 images featuring two-character alphanumeric strings embedded in Ishihara test style dot patterns, challenging models on precise pattern recognition. Our evaluation of nine state-of-the-art MLLMs on HueManity demonstrates a significant performance deficit compared to human and traditional computer vision baselines. The best-performing MLLM achieved a 33.6% accuracy on the numeric `easy' task and a striking 3% on the alphanumeric `hard' task. In contrast, human participants achieved near-perfect scores (100% and 95.6%), and a fine-tuned ResNet50 model reached accuracies of 96.5% and 94.5%. These results highlight a critical gap in the visual capabilities of current MLLMs. Our analysis further explores potential architectural and training-paradigm factors contributing to this perceptual gap in MLLMs. We open-source HueManity dataset and code to foster further research in improving perceptual robustness of MLLMs. | [
{
"version": "v1",
"created": "Sat, 31 May 2025 22:59:48 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Grover",
"Rynaa",
""
],
[
"Tamarapalli",
"Jayant Sravan",
""
],
[
"Yerramilli",
"Sahiti",
""
],
[
"Pande",
"Nilay",
""
]
] |
2506.03195 | Yunqi Hong | Yunqi Hong, Sohyun An, Andrew Bai, Neil Y.C. Lin and Cho-Jui Hsieh | Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs | null | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite Multimodal Large Language Models (MLLMs) showing promising results on general zero-shot image classification tasks, fine-grained image classification remains challenging. It demands precise attention to subtle visual details to distinguish between visually similar subcategories--details that MLLMs may easily overlook without explicit guidance. To address this, we introduce AutoSEP, an iterative self-supervised prompt learning framework designed to enhance MLLM fine-grained classification capabilities in a fully unsupervised manner. Our core idea is to leverage unlabeled data to learn a description prompt that guides MLLMs in identifying crucial discriminative features within an image, and boosts classification accuracy. We developed an automatic self-enhancing prompt learning framework called AutoSEP to iteratively improve the description prompt using unlabeled data, based on instance-level classification scoring function. AutoSEP only requires black-box access to MLLMs, eliminating the need for any training or fine-tuning. We evaluate our approach on multiple fine-grained classification datasets. It consistently outperforms other unsupervised baselines, demonstrating the effectiveness of our self-supervised optimization framework. Notably, AutoSEP on average improves 13 percent over standard zero-shot classification and 5 percent over the best-performing baselines. Code is available at: https://github.com/yq-hong/AutoSEP | [
{
"version": "v1",
"created": "Sun, 01 Jun 2025 09:04:07 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Hong",
"Yunqi",
""
],
[
"An",
"Sohyun",
""
],
[
"Bai",
"Andrew",
""
],
[
"Lin",
"Neil Y. C.",
""
],
[
"Hsieh",
"Cho-Jui",
""
]
] |
2506.03196 | Dania Herzalla | Dania Herzalla, Willian T. Lunardi and Martin Andreoni | Graph Neural Networks for Jamming Source Localization | null | null | null | null | cs.NI cs.CR cs.IT cs.LG eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph-based learning has emerged as a transformative approach for modeling complex relationships across diverse domains, yet its potential in wireless security remains largely unexplored. In this work, we introduce the first application of graph-based learning for jamming source localization, addressing the imminent threat of jamming attacks in wireless networks. Unlike geometric optimization techniques that struggle under environmental uncertainties and dense interference, we reformulate localization as an inductive graph regression task. Our approach integrates structured node representations that encode local and global signal aggregation, ensuring spatial coherence and adaptive signal fusion. To enhance robustness, we incorporate an attention-based graph neural network that adaptively refines neighborhood influence and introduces a confidence-guided estimation mechanism that dynamically balances learned predictions with domain-informed priors. We evaluate our approach under complex radio frequency environments with varying sampling densities and signal propagation conditions, conducting comprehensive ablation studies on graph construction, feature selection, and pooling strategies. Results demonstrate that our novel graph-based learning framework significantly outperforms established localization baselines, particularly in challenging scenarios with sparse and obfuscated signal information. Code is available at [https://github.com/daniaherzalla/gnn-jamming-source-localization](https://github.com/daniaherzalla/gnn-jamming-source-localization). | [
{
"version": "v1",
"created": "Sun, 01 Jun 2025 14:29:25 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Herzalla",
"Dania",
""
],
[
"Lunardi",
"Willian T.",
""
],
[
"Andreoni",
"Martin",
""
]
] |
2506.03197 | Biao Wu | Baode Wang, Biao Wu, Weizhen Li, Meng Fang, Yanjie Liang, Zuming Huang, Haozhe Wang, Jun Huang, Ling Chen, Wei Chu, Yuan Qi | Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing | 16 pages, 12 figures | null | null | INF-CS-TR-2025-02 | cs.CV cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automated parsing of scanned documents into richly structured, machine-readable formats remains a critical bottleneck in Document AI, as traditional multi-stage pipelines suffer from error propagation and limited adaptability to diverse layouts. We introduce layoutRL, an end-to-end reinforcement learning framework that trains models to be explicitly layout-aware by optimizing a composite reward of normalized edit distance, paragraph count accuracy, and reading order preservation. Leveraging our newly released dataset, Infinity-Doc-55K, which combines 55K high-fidelity synthetic scanned document parsing data with expert-filtered real-world documents, we instantiate layoutRL in a vision-language-model-based parser called Infinity-Parser. Evaluated on English and Chinese benchmarks for OCR, table and formula extraction, and reading order detection, Infinity-Parser achieves new state-of-the-art performance in both accuracy and structural fidelity, outpacing specialist pipelines and general-purpose vision-language models. We will publicly release our code and dataset to accelerate progress in robust document understanding. | [
{
"version": "v1",
"created": "Sun, 01 Jun 2025 15:19:52 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Wang",
"Baode",
""
],
[
"Wu",
"Biao",
""
],
[
"Li",
"Weizhen",
""
],
[
"Fang",
"Meng",
""
],
[
"Liang",
"Yanjie",
""
],
[
"Huang",
"Zuming",
""
],
[
"Wang",
"Haozhe",
""
],
[
"Huang",
"Jun",
""
],
[
"Chen",
"Ling",
""
],
[
"Chu",
"Wei",
""
],
[
"Qi",
"Yuan",
""
]
] |
2506.03198 | Hao Yin | Hao Yin, Lijun Gu, Paritosh Parmar, Lin Xu, Tianxiao Guo, Weiwei Fu, Yang Zhang, Tianyou Zheng | FLEX: A Large-Scale Multi-Modal Multi-Action Dataset for Fitness Action Quality Assessment | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | With the increasing awareness of health and the growing desire for aesthetic physique, fitness has become a prevailing trend. However, the potential risks associated with fitness training, especially with weight-loaded fitness actions, cannot be overlooked. Action Quality Assessment (AQA), a technology that quantifies the quality of human action and provides feedback, holds the potential to assist fitness enthusiasts of varying skill levels in achieving better training outcomes. Nevertheless, current AQA methodologies and datasets are limited to single-view competitive sports scenarios and RGB modality and lack professional assessment and guidance of fitness actions. To address this gap, we propose the FLEX dataset, the first multi-modal, multi-action, large-scale dataset that incorporates surface electromyography (sEMG) signals into AQA. FLEX utilizes high-precision MoCap to collect 20 different weight-loaded actions performed by 38 subjects across 3 different skill levels for 10 repetitions each, containing 5 different views of the RGB video, 3D pose, sEMG, and physiological information. Additionally, FLEX incorporates knowledge graphs into AQA, constructing annotation rules in the form of penalty functions that map weight-loaded actions, action keysteps, error types, and feedback. We conducted various baseline methodologies on FLEX, demonstrating that multimodal data, multiview data, and fine-grained annotations significantly enhance model performance. FLEX not only advances AQA methodologies and datasets towards multi-modal and multi-action scenarios but also fosters the integration of artificial intelligence within the fitness domain. Dataset and code are available at https://haoyin116.github.io/FLEX_Dataset. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 01:44:02 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Yin",
"Hao",
""
],
[
"Gu",
"Lijun",
""
],
[
"Parmar",
"Paritosh",
""
],
[
"Xu",
"Lin",
""
],
[
"Guo",
"Tianxiao",
""
],
[
"Fu",
"Weiwei",
""
],
[
"Zhang",
"Yang",
""
],
[
"Zheng",
"Tianyou",
""
]
] |
2506.03199 | Luca Candelori | Giuseppe Di Caro, Vahagn Kirakosyan, Alexander G. Abanov, Luca Candelori, Nadine Hartmann, Ernest T. Lam, Kharen Musaelian, Ryan Samson, Dario Villani, Martin T. Wells, Richard J. Wenstrup, Mengjia Xu | Quantum Cognition Machine Learning for Forecasting Chromosomal Instability | null | null | null | null | q-bio.QM cs.LG quant-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The accurate prediction of chromosomal instability from the morphology of circulating tumor cells (CTCs) enables real-time detection of CTCs with high metastatic potential in the context of liquid biopsy diagnostics. However, it presents a significant challenge due to the high dimensionality and complexity of single-cell digital pathology data. Here, we introduce the application of Quantum Cognition Machine Learning (QCML), a quantum-inspired computational framework, to estimate morphology-predicted chromosomal instability in CTCs from patients with metastatic breast cancer. QCML leverages quantum mechanical principles to represent data as state vectors in a Hilbert space, enabling context-aware feature modeling, dimensionality reduction, and enhanced generalization without requiring curated feature selection. QCML outperforms conventional machine learning methods when tested on out of sample verification CTCs, achieving higher accuracy in identifying predicted large-scale state transitions (pLST) status from CTC-derived morphology features. These preliminary findings support the application of QCML as a novel machine learning tool with superior performance in high-dimensional, low-sample-size biomedical contexts. QCML enables the simulation of cognition-like learning for the identification of biologically meaningful prediction of chromosomal instability from CTC morphology, offering a novel tool for CTC classification in liquid biopsy. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 13:55:33 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Di Caro",
"Giuseppe",
""
],
[
"Kirakosyan",
"Vahagn",
""
],
[
"Abanov",
"Alexander G.",
""
],
[
"Candelori",
"Luca",
""
],
[
"Hartmann",
"Nadine",
""
],
[
"Lam",
"Ernest T.",
""
],
[
"Musaelian",
"Kharen",
""
],
[
"Samson",
"Ryan",
""
],
[
"Villani",
"Dario",
""
],
[
"Wells",
"Martin T.",
""
],
[
"Wenstrup",
"Richard J.",
""
],
[
"Xu",
"Mengjia",
""
]
] |
2506.03200 | Paul Sokol | Stephanie McNamara, Prabin Parajuli, Sutirtha Paul, Garfield Warren, Adrian Del Maestro, Paul E. Sokol | Novel Experimental Platform to realize One-dimensional Quantum Fluids | 10 pages, 3 figures | null | null | null | cond-mat.mtrl-sci cond-mat.quant-gas | http://creativecommons.org/licenses/by/4.0/ | Templated porous materials, such as MCM-41, due to the uniformity of their onedimensional structure and scalability in synthesis, have emerged as an attractive medium for studying one-dimensional quantum fluids. However, the experimental challenge of synthesizing these materials with pore radii smaller than 15 Angstroms hinders the realization of a one-dimensional quantum liquid of helium within such systems, as the coherence length of helium is shorter than the pore radius. Recently, DelMaestro et. al. have preplated MCM-41 with Ar resulting in a reduction of the pore size and a softening of the adsorption potential allowing them to observe 1D Tomanga-Luttinger liquid like behavior. In this paper we present a novel method to obtain an even more ideal environment for studying the behavior of 1D 4He. We propose preplating MCM-41 pores with cesium (Cs) metal. The non-wetting nature of helium on a Cs-coated surface, coupled with the large atomic radius of cesium, creates an optimal environment for confining a quantum liquid of helium in one-dimensional geometry. We present preliminary measurements of adsorption isotherms and Small Angle X-ray Scattering studies that 1reveal a reduction in pore radius upon preplating MCM-41 with Cs, demonstrating promising prospects for facilitating the realization of one-dimensional quantum fluids in templated porous materials. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 14:56:30 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"McNamara",
"Stephanie",
""
],
[
"Parajuli",
"Prabin",
""
],
[
"Paul",
"Sutirtha",
""
],
[
"Warren",
"Garfield",
""
],
[
"Del Maestro",
"Adrian",
""
],
[
"Sokol",
"Paul E.",
""
]
] |
2506.03201 | Joseph Paddison | Joseph A. M. Paddison, Stuart Calder, Danielle R. Yahne, Malcolm J. Cochran, Si Athena Chen, Matthias D. Frontzek, and Yuanpeng Zhang | A planning tool for neutron powder diffraction experiments | 8 pages, 3 figures | null | null | null | physics.ins-det cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a computer program to simulate the results of neutron powder-diffraction experiments at the High Flux Isotope Reactor at Oak Ridge National Laboratory. The program is freely available as a web application at http://addie.ornl.gov/hfirestimate, and is designed to be straightforward to use for researchers who are new to neutron diffraction. The input includes the crystal structure of the proposed sample, the sample mass, and the instrument configuration. The results include a plot of the simulated data -- including realistic estimates of background and the error bars due to counting statistics -- and suggestions of how to resolve potential problems with the experiment. Here, we explain the design and implementation of this program and demonstrate its performance using comparisons of simulated and experimental data. We hope that this program will enable researchers to plan neutron-scattering experiments more effectively, increasing the likelihood of successful experiments and improving the productivity of neutron-diffraction research. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 15:08:37 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Paddison",
"Joseph A. M.",
""
],
[
"Calder",
"Stuart",
""
],
[
"Yahne",
"Danielle R.",
""
],
[
"Cochran",
"Malcolm J.",
""
],
[
"Chen",
"Si Athena",
""
],
[
"Frontzek",
"Matthias D.",
""
],
[
"Zhang",
"Yuanpeng",
""
]
] |
2506.03202 | Itxasne Ant\'unez S\'aenz | Itxasne Ant\'unez S\'aenz, Ane Alberdi Aramendi, David Dunaway, Juling Ong, Lara Deli\`ege, Amparo S\'aenz, Anita Ahmadi Birjandi, Noor UI Owase Jeelani, Silvia Schievano and Alessandro Borghi | A combined Machine Learning and Finite Element Modelling tool for the surgical planning of craniosynostosis correction | 11 pages, 16 figures | null | null | null | eess.IV cs.CV cs.LG physics.med-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Craniosynostosis is a medical condition that affects the growth of babies' heads, caused by an early fusion of cranial sutures. In recent decades, surgical treatments for craniosynostosis have significantly improved, leading to reduced invasiveness, faster recovery, and less blood loss. At Great Ormond Street Hospital (GOSH), the main surgical treatment for patients diagnosed with sagittal craniosynostosis (SC) is spring assisted cranioplasty (SAC). This procedure involves a 15x15 mm2 osteotomy, where two springs are inserted to induce distraction. Despite the numerous advantages of this surgical technique for patients, the outcome remains unpredictable due to the lack of efficient preoperative planning tools. The surgeon's experience and the baby's age are currently relied upon to determine the osteotomy location and spring selection. Previous tools for predicting the surgical outcome of SC relied on finite element modeling (FEM), which involved computed tomography (CT) imaging and required engineering expertise and lengthy calculations. The main goal of this research is to develop a real-time prediction tool for the surgical outcome of patients, eliminating the need for CT scans to minimise radiation exposure during preoperative planning. The proposed methodology involves creating personalised synthetic skulls based on three-dimensional (3D) photographs, incorporating population average values of suture location, skull thickness, and soft tissue properties. A machine learning (ML) surrogate model is employed to achieve the desired surgical outcome. The resulting multi-output support vector regressor model achieves a R2 metric of 0.95 and MSE and MAE below 0.13. Furthermore, in the future, this model could not only simulate various surgical scenarios but also provide optimal parameters for achieving a maximum cranial index (CI). | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 17:11:50 GMT"
}
] | 2025-06-05T00:00:00 | [
[
"Sáenz",
"Itxasne Antúnez",
""
],
[
"Aramendi",
"Ane Alberdi",
""
],
[
"Dunaway",
"David",
""
],
[
"Ong",
"Juling",
""
],
[
"Deliège",
"Lara",
""
],
[
"Sáenz",
"Amparo",
""
],
[
"Birjandi",
"Anita Ahmadi",
""
],
[
"Jeelani",
"Noor UI Owase",
""
],
[
"Schievano",
"Silvia",
""
],
[
"Borghi",
"Alessandro",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.