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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.08578 | Boyang Chen | Boyang Chen, Xizhe Zang, Chao Song, Yue Zhang, Xuehe Zhang and Jie Zhao | Noise Analysis and Hierarchical Adaptive Body State Estimator For Biped Robot Walking With ESVC Foot | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The ESVC(Ellipse-based Segmental Varying Curvature) foot, a robot foot design inspired by the rollover shape of the human foot, significantly enhances the energy efficiency of the robot walking gait. However, due to the tilt of the supporting leg, the error of the contact model are amplified, making robot state estimation more challenging. Therefore, this paper focuses on the noise analysis and state estimation for robot walking with the ESVC foot. First, through physical robot experiments, we investigate the effect of the ESVC foot on robot measurement noise and process noise. and a noise-time regression model using sliding window strategy is developed. Then, a hierarchical adaptive state estimator for biped robots with the ESVC foot is proposed. The state estimator consists of two stages: pre-estimation and post-estimation. In the pre-estimation stage, a data fusion-based estimation is employed to process the sensory data. During post-estimation, the acceleration of center of mass is first estimated, and then the noise covariance matrices are adjusted based on the regression model. Following that, an EKF(Extended Kalman Filter) based approach is applied to estimate the centroid state during robot walking. Physical experiments demonstrate that the proposed adaptive state estimator for biped robot walking with the ESVC foot not only provides higher precision than both EKF and Adaptive EKF, but also converges faster under varying noise conditions. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:48:20 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Chen",
"Boyang",
""
],
[
"Zang",
"Xizhe",
""
],
[
"Song",
"Chao",
""
],
[
"Zhang",
"Yue",
""
],
[
"Zhang",
"Xuehe",
""
],
[
"Zhao",
"Jie",
""
]
] |
2506.08580 | Lv Yang | Yang Lv, Jinlong Lei and Peng Yi | HGFormer: A Hierarchical Graph Transformer Framework for Two-Stage Colonel Blotto Games via Reinforcement Learning | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Two-stage Colonel Blotto game represents a typical adversarial resource allocation problem, in which two opposing agents sequentially allocate resources in a network topology across two phases: an initial resource deployment followed by multiple rounds of dynamic reallocation adjustments. The sequential dependency between game stages and the complex constraints imposed by the graph topology make it difficult for traditional approaches to attain a globally optimal strategy. To address these challenges, we propose a hierarchical graph Transformer framework called HGformer. By incorporating an enhanced graph Transformer encoder with structural biases and a two-agent hierarchical decision model, our approach enables efficient policy generation in large-scale adversarial environments. Moreover, we design a layer-by-layer feedback reinforcement learning algorithm that feeds the long-term returns from lower-level decisions back into the optimization of the higher-level strategy, thus bridging the coordination gap between the two decision-making stages. Experimental results demonstrate that, compared to existing hierarchical decision-making or graph neural network methods, HGformer significantly improves resource allocation efficiency and adversarial payoff, achieving superior overall performance in complex dynamic game scenarios. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:51:18 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Lv",
"Yang",
""
],
[
"Lei",
"Jinlong",
""
],
[
"Yi",
"Peng",
""
]
] |
2506.08583 | Nicolas Chevalier | Nicolas R. Chevalier, Alexis Peaucelle (IJPB), Thomas Guilbert (IC UM3), Pierre Bourdoncle, Wang Xi | The enteric nervous system is 10 times stiffer than the brain | null | Biophysical Journal, 2025 | null | null | physics.bio-ph q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural tissues of the central nervous system are among the softest and most fragile in the human body, protected from mechanical perturbation by the skull and the spine. In contrast, the enteric nervous system is embedded in a compliant, contractile tissue and subject to chronic, high-magnitude mechanical stress. Do neurons and glia of the enteric nervous system display specific mechanical properties to withstand these forces? Using nano-indentation combined with immunohistochemistry and second harmonic generation imaging of collagen, we discovered that enteric ganglia in adult mice are an order of magnitude more resistant to deformation than brain tissue. We found that glia-rich regions in ganglia have a similar stiffness to neuron-rich regions and to the surrounding smooth muscle, of ~3 kPa at 3 $\mu$m indentation depth and of ~7 kPa at 8 $\mu$m depth. Differences in the adhesion strength of the different tissue layers to the glass indenter were scarce. The collagen shell surrounding ganglia and inter-ganglionic fibers may play a key role in strengthening the enteric nervous system to resist the manifold mechanical challenges it faces. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:52:31 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Chevalier",
"Nicolas R.",
"",
"IJPB"
],
[
"Peaucelle",
"Alexis",
"",
"IJPB"
],
[
"Guilbert",
"Thomas",
"",
"IC UM3"
],
[
"Bourdoncle",
"Pierre",
""
],
[
"Xi",
"Wang",
""
]
] |
2506.08584 | Yahan Li | Yahan Li, Jifan Yao, John Bosco S. Bunyi, Adam C. Frank, Angel Hwang, Ruishan Liu | CounselBench: A Large-Scale Expert Evaluation and Adversarial Benchmark of Large Language Models in Mental Health Counseling | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Large language models (LLMs) are increasingly proposed for use in mental health support, yet their behavior in realistic counseling scenarios remains largely untested. We introduce CounselBench, a large-scale benchmark developed with 100 mental health professionals to evaluate and stress-test LLMs in single-turn counseling. The first component, CounselBench-EVAL, contains 2,000 expert evaluations of responses from GPT-4, LLaMA 3, Gemini, and online human therapists to real patient questions. Each response is rated along six clinically grounded dimensions, with written rationales and span-level annotations. We find that LLMs often outperform online human therapists in perceived quality, but experts frequently flag their outputs for safety concerns such as unauthorized medical advice. Follow-up experiments show that LLM judges consistently overrate model responses and overlook safety issues identified by human experts. To probe failure modes more directly, we construct CounselBench-Adv, an adversarial dataset of 120 expert-authored counseling questions designed to trigger specific model issues. Evaluation across 2,880 responses from eight LLMs reveals consistent, model-specific failure patterns. Together, CounselBench establishes a clinically grounded framework for benchmarking and improving LLM behavior in high-stakes mental health settings. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:53:06 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Li",
"Yahan",
""
],
[
"Yao",
"Jifan",
""
],
[
"Bunyi",
"John Bosco S.",
""
],
[
"Frank",
"Adam C.",
""
],
[
"Hwang",
"Angel",
""
],
[
"Liu",
"Ruishan",
""
]
] |
2506.08585 | Petr Hlin\v{e}n\'y | Petr Hlin\v{e}n\'y and Jan Jedelsk\'y | k-Planar and Fan-Crossing Drawings and Transductions of Planar Graphs | null | null | null | null | cs.CG cs.LO math.CO | http://creativecommons.org/licenses/by-sa/4.0/ | We introduce a two-way connection between FO transductions (logical transformations) of planar graphs, and a certain variant of fan-crossing (fan-planar) drawings of graphs which for bounded-degree graphs essentially reduces to being k-planar for fixed k. For graph classes, this connection allows to derive non-transducibility results from nonexistence of the said drawings and, conversely, from nonexistence of a transduction to derive nonexistence of the said drawings. For example, the class of 3D-grids is not k-planar for any fixed k. We hope that this connection will help to draw a path to a possible proof that not all toroidal graphs are transducible from planar graphs.
Our characterization can be extended to any fixed surface instead of the plane. The result is based on a very recent characterization of weakly sparse FO transductions of classes of bounded expansion by [Gajarsk\'y, G{\l}adkowski, Jedelsk\'y, Pilipczuk and Toru\'nczyk, arXiv:2505.15655]. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:54:08 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Hliněný",
"Petr",
""
],
[
"Jedelský",
"Jan",
""
]
] |
2506.08586 | Gildas Mazo | Ekaterina Tomilina (MaIAGE, GABI), Florence Jaffr\'ezic (GABI), Gildas Mazo (MaIAGE) | Gaussian copula correlation network analysis with application to multi-omics data | null | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reconstructing gene regulatory networks from large-scale heterogeneous data is a key challenge in biology. In multi-omics data analysis, networks based on pairwise statistical association measures remain popular, as they are easy to build and understand. In the presence of mixed-type (discrete and continuous) data, however, the choice of good association measures remains an important issue. We propose here a novel approach based on the Gaussian copula, the parameters of which represent the links of the network. Novel properties of the model are obtained to guide the interpretation of the network. To estimate the copula parameters, we calculated a semiparametric pairwise likelihood for mixed data. In an extensive simulation study, we showed that the proposed estimation procedure was able to accurately estimate the copula correlation matrix. The proposed methodology was also applied to a real ICGC dataset on breast cancer, and is implemented in a freely available R package heterocop. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:55:20 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Tomilina",
"Ekaterina",
"",
"MaIAGE, GABI"
],
[
"Jaffrézic",
"Florence",
"",
"GABI"
],
[
"Mazo",
"Gildas",
"",
"MaIAGE"
]
] |
2506.08587 | Ziqian Hua | Ziqian Hua, Zhiyuan Li | Chandra X-ray Measurement of Heavy Element Abundances of Wolf-Rayet Stars in the Galactic Center | 13 pages, 7 figures, Accepted for publication in MNRAS | null | null | null | astro-ph.GA astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | Elemental abundances hold important information about the star formation history in the Galactic Center. The thermal X-ray spectra of certain stars can provide a robust probe of elemental abundances, mainly through the presence of K-shell emission lines. In this work, based on deep archival {\it Chandra} observations, we obtain X-ray measurements of five heavy elements (Si, S, Ar, Ca and Fe) for three sources in the Arches cluster, one source in the Quintuplet cluster, as well as a field source known as Edd 1, which are all probable WR stars exhibiting a high quality X-ray spectrum. A two-temperature, non-equilibrium ionization plasma model is employed for the spectral fit, taking into account light element compositions characteristic of WR star winds, which is substantially depleted in hydrogen but enriched in nitrogen and/or carbon. It is found that the Arches and Quintuplet WR stars share similar abundances of Si, S, and Ar, while exhibiting distinct Ca and Fe abundances, which may be understood as due to dust depletion of the latter two elements in Quintuplet. The observed near-solar or sub-solar metallicity of the WR star winds can be naturally understood as the result of nucleosynthesis and internal mixing of the parent star, which have a supersolar initial metallicity as expected for the Galactic center in general. Implications of our findings on the origin of the young star clusters and isolated massive stars in the Galactic center, as well as the elemental composition of the accretion flow onto Sgr A*, are addressed. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:55:52 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Hua",
"Ziqian",
""
],
[
"Li",
"Zhiyuan",
""
]
] |
2506.08588 | Eugenio G. Omodeo | Liesbeth De Mol, Yuri V. Matiyasevich, Eugenio G. Omodeo, Alberto Policriti, Wilfried Sieg, Elaine J. Weyuker | Martin Davis: An Overview of his Work in Logic, Computer Science, and Philosophy | null | null | null | null | cs.LO | http://creativecommons.org/licenses/by-sa/4.0/ | In his autobiographic essay written in 1999, ``From logic to computer science and back'', Martin David Davis (3/8/1928--1/1/2023) indicated that he viewed himself as a logician \emph{and} a computer scientist. He expanded the essay in 2016 and expressed a new perspective through a changed title, ``My life as a logician''. He points out that logic was the unifying theme underlying his scientific career. Our paper attempts to provide a consistent vision that illuminates Davis' successive contributions leading to his landmark writings on computability, unsolvable problems, automated reasoning, as well as the history and philosophy of computing. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:56:48 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"De Mol",
"Liesbeth",
""
],
[
"Matiyasevich",
"Yuri V.",
""
],
[
"Omodeo",
"Eugenio G.",
""
],
[
"Policriti",
"Alberto",
""
],
[
"Sieg",
"Wilfried",
""
],
[
"Weyuker",
"Elaine J.",
""
]
] |
2506.08589 | Huseyin Dag | \"O. E. Demircan, H. Da\u{g}, H. Sundu, J. Y. S\"ung\"u and E. V. Veliev | An interpretation of the fully-charmed scalar state $X(6200)$ as a molecular di-meson | 15 pages, 4 figures | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | The LHCb Collaboration recently observed new structures in the invariant mass spectrum of $J/\psi J/\psi$ meson pairs produced in proton-proton collisions, including a narrow peak around $6.2$ GeV. This study investigates the thermal behavior of this newly discovered fully-charm resonance, assuming it to be an $\eta_c\eta_c$ molecule with quantum numbers $J^{PC} = 0^{++}$. Employing thermal QCD sum rules up to dimension four, we analyze the temperature dependence of the mass and the decay constant. Our results indicate that both physical quantities decrease as the temperature rises. Notably, at zero temperature, the mass and decay constant of the state are consistent with those reported in the existing literature. These findings are expected to provide valuable insights for future experimental investigations in this field. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:57:50 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Demircan",
"Ö. E.",
""
],
[
"Dağ",
"H.",
""
],
[
"Sundu",
"H.",
""
],
[
"Süngü",
"J. Y.",
""
],
[
"Veliev",
"E. V.",
""
]
] |
2506.08590 | Thomas Gamet | Thomas Gamet (UMPA-ENSL) | Renormalization of bosonic quadratic Hamiltonians involving rank-one perturbations | null | null | null | null | math-ph math.MP math.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this article, we study the renormalization of a bosonic quadratic Hamiltonian with an ultraviolet divergence. The Hamiltonian is composed of the sum of a free energy and the square of the smeared field operator. We explicitly diagonalize the Hamiltonian via Bogoliubov transformations, and the diagonalized Hamiltonian is easier to study. We show a renormalization of both the energy and the coupling constant depending on the form factor of the interaction. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:59:11 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Gamet",
"Thomas",
"",
"UMPA-ENSL"
]
] |
2506.08591 | Chengchao Shen | Chengchao Shen, Hourun Zhu, Gongfan Fang, Jianxin Wang, Xinchao Wang | Diversity-Guided MLP Reduction for Efficient Large Vision Transformers | null | null | null | null | cs.CV cs.LG cs.MM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Transformer models achieve excellent scaling property, where the performance is improved with the increment of model capacity. However, large-scale model parameters lead to an unaffordable cost of computing and memory. We analyze popular transformer architectures and find that multilayer perceptron (MLP) modules take up the majority of model parameters.
To this end, we focus on the recoverability of the compressed models and propose a Diversity-Guided MLP Reduction (DGMR) method to significantly reduce the parameters of large vision transformers with only negligible performance degradation. Specifically, we conduct a Gram-Schmidt weight pruning strategy to eliminate redundant neurons of MLP hidden layer, while preserving weight diversity for better performance recover during distillation. Compared to the model trained from scratch, our pruned model only requires 0.06\% data of LAION-2B (for the training of large vision transformers) without labels (ImageNet-1K) to recover the original performance. Experimental results on several state-of-the-art large vision transformers demonstrate that our method achieves a more than 57.0\% parameter and FLOPs reduction in a near lossless manner. Notably, for EVA-CLIP-E (4.4B), our method accomplishes a 71.5\% parameter and FLOPs reduction without performance degradation. The source code and trained weights are available at https://github.com/visresearch/DGMR. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 08:59:27 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Shen",
"Chengchao",
""
],
[
"Zhu",
"Hourun",
""
],
[
"Fang",
"Gongfan",
""
],
[
"Wang",
"Jianxin",
""
],
[
"Wang",
"Xinchao",
""
]
] |
2506.08592 | Liyan Xu | Liyan Xu, Zhenlin Su, Mo Yu, Jiangnan Li, Fandong Meng, Jie Zhou | Dense Retrievers Can Fail on Simple Queries: Revealing The Granularity Dilemma of Embeddings | null | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work focuses on an observed limitation of text encoders: embeddings may not be able to recognize fine-grained entities or events within the semantics, resulting in failed dense retrieval on even simple cases. To examine such behaviors, we first introduce a new evaluation dataset in Chinese, named CapRetrieval, whose passages are image captions, and queries are phrases inquiring entities or events in various forms. Zero-shot evaluation suggests that encoders may fail on these fine-grained matching, regardless of training sources or model sizes. Aiming for enhancement, we proceed to finetune encoders with our proposed data generation strategies, which obtains the best performance on CapRetrieval. Within this process, we further identify an issue of granularity dilemma, a challenge for embeddings to express fine-grained salience while aligning with overall semantics. Our dataset, code and models in this work are publicly released at https://github.com/lxucs/CapRetrieval. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:00:33 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Xu",
"Liyan",
""
],
[
"Su",
"Zhenlin",
""
],
[
"Yu",
"Mo",
""
],
[
"Li",
"Jiangnan",
""
],
[
"Meng",
"Fandong",
""
],
[
"Zhou",
"Jie",
""
]
] |
2506.08593 | Mario Tawfelis | Shuzhou Yuan, Ercong Nie, Mario Tawfelis, Helmut Schmid, Hinrich Sch\"utze, and Michael F\"arber | Hateful Person or Hateful Model? Investigating the Role of Personas in Hate Speech Detection by Large Language Models | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Hate speech detection is a socially sensitive and inherently subjective task, with judgments often varying based on personal traits. While prior work has examined how socio-demographic factors influence annotation, the impact of personality traits on Large Language Models (LLMs) remains largely unexplored. In this paper, we present the first comprehensive study on the role of persona prompts in hate speech classification, focusing on MBTI-based traits. A human annotation survey confirms that MBTI dimensions significantly affect labeling behavior. Extending this to LLMs, we prompt four open-source models with MBTI personas and evaluate their outputs across three hate speech datasets. Our analysis uncovers substantial persona-driven variation, including inconsistencies with ground truth, inter-persona disagreement, and logit-level biases. These findings highlight the need to carefully define persona prompts in LLM-based annotation workflows, with implications for fairness and alignment with human values. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:02:55 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Yuan",
"Shuzhou",
""
],
[
"Nie",
"Ercong",
""
],
[
"Tawfelis",
"Mario",
""
],
[
"Schmid",
"Helmut",
""
],
[
"Schütze",
"Hinrich",
""
],
[
"Färber",
"Michael",
""
]
] |
2506.08594 | Weikang Li | Yixuan Ma, Chang Liu, Weikang Li, Shun-Yao Zhang, L.-M. Duan, Yukai Wu, Dong-Ling Deng | Solving excited states for long-range interacting trapped ions with neural networks | null | null | null | null | quant-ph cond-mat.dis-nn cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The computation of excited states in strongly interacting quantum many-body systems is of fundamental importance. Yet, it is notoriously challenging due to the exponential scaling of the Hilbert space dimension with the system size. Here, we introduce a neural network-based algorithm that can simultaneously output multiple low-lying excited states of a quantum many-body spin system in an accurate and efficient fashion. This algorithm, dubbed the neural quantum excited-state (NQES) algorithm, requires no explicit orthogonalization of the states and is generally applicable to higher dimensions. We demonstrate, through concrete examples including the Haldane-Shastry model with all-to-all interactions, that the NQES algorithm is capable of efficiently computing multiple excited states and their related observable expectations. In addition, we apply the NQES algorithm to two classes of long-range interacting trapped-ion systems in a two-dimensional Wigner crystal. For non-decaying all-to-all interactions with alternating signs, our computed low-lying excited states bear spatial correlation patterns similar to those of the ground states, which closely match recent experimental observations that the quasi-adiabatically prepared state accurately reproduces analytical ground-state correlations. For a system of up to 300 ions with power-law decaying antiferromagnetic interactions, we successfully uncover its gap scaling and correlation features. Our results establish a scalable and efficient algorithm for computing excited states of interacting quantum many-body systems, which holds potential applications ranging from benchmarking quantum devices to photoisomerization. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:02:59 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Ma",
"Yixuan",
""
],
[
"Liu",
"Chang",
""
],
[
"Li",
"Weikang",
""
],
[
"Zhang",
"Shun-Yao",
""
],
[
"Duan",
"L. -M.",
""
],
[
"Wu",
"Yukai",
""
],
[
"Deng",
"Dong-Ling",
""
]
] |
2506.08596 | Guyang Zhang | Guyang Zhang, Waleed Abdulla | Transformers Meet Hyperspectral Imaging: A Comprehensive Study of Models, Challenges and Open Problems | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformers have become the architecture of choice for learning long-range dependencies, yet their adoption in hyperspectral imaging (HSI) is still emerging. We reviewed more than 300 papers published up to 2025 and present the first end-to-end survey dedicated to Transformer-based HSI classification. The study categorizes every stage of a typical pipeline-pre-processing, patch or pixel tokenization, positional encoding, spatial-spectral feature extraction, multi-head self-attention variants, skip connections, and loss design-and contrasts alternative design choices with the unique spatial-spectral properties of HSI. We map the field's progress against persistent obstacles: scarce labeled data, extreme spectral dimensionality, computational overhead, and limited model explainability. Finally, we outline a research agenda prioritizing valuable public data sets, lightweight on-edge models, illumination and sensor shifts robustness, and intrinsically interpretable attention mechanisms. Our goal is to guide researchers in selecting, combining, or extending Transformer components that are truly fit for purpose for next-generation HSI applications. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:04:30 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Zhang",
"Guyang",
""
],
[
"Abdulla",
"Waleed",
""
]
] |
2506.08597 | Hamid Omidi | H. Omidi, L. Sacco, V. Hutter, G. Irsiegler, M. Claus, M. Schobben, A. Jacob, M. Schramm, S. Fiore | Towards Provenance-Aware Earth Observation Workflows: the openEO Case Study | null | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Capturing the history of operations and activities during a computational workflow is significantly important for Earth Observation (EO). The data provenance helps to collect the metadata that records the lineage of data products, providing information about how data are generated, transferred, manipulated, by whom all these operations are performed and through which processes, parameters, and datasets. This paper presents an approach to improve those aspects, by integrating the data provenance library yProv4WFs within openEO, a platform to let users connect to Earth Observation cloud back-ends in a simple and unified way. In addition, it is demonstrated how the integration of data provenance concepts across EO processing chains enables researchers and stakeholders to better understand the flow, the dependencies, and the transformations involved in analytical workflows. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:05:00 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Omidi",
"H.",
""
],
[
"Sacco",
"L.",
""
],
[
"Hutter",
"V.",
""
],
[
"Irsiegler",
"G.",
""
],
[
"Claus",
"M.",
""
],
[
"Schobben",
"M.",
""
],
[
"Jacob",
"A.",
""
],
[
"Schramm",
"M.",
""
],
[
"Fiore",
"S.",
""
]
] |
2506.08598 | Gleb Pokrovski | Gleb S. Pokrovski (GET) | Quantifying sulfur speciation in magmatic-hydrothermal fluids | null | null | null | null | physics.geo-ph physics.chem-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantitative knowledge of sulfur speciation in the fluid phase is key to understanding sulfur degassing from magmas and its transfer along with metals by fluids across the lithosphere. Farsang and Zajacz (1) recently reported new sulfur speciation data using Raman spectroscopy measurements on aqueous H$_2$SO$_4$-NaCl(-KCl) solutions trapped as synthetic fluid inclusions in quartz at 875$^\circ$C and 2 kbar under variable redox conditions. They interpreted the data by dominant SO$_{2(aq)}$ and HS$^-$, along with subordinate H$_2$S$_{(aq)}$, whereas the di-and trisulfur radical ions, $[S_3^\bullet]^-$ and $[S_2^\bullet]^-$, reported in aqueous fluids both in nature and experiment to at least 700 $^\circ$C and 15 kbar (2-5) , were undetectable in Raman nonresonant spectra 1 . On the basis of thermochemical calculations that returned negligible HS$^-$ and large $[S_3^\bullet]^-$ concentrations at their experimental conditions, the authors claimed that the published $[S_3^\bullet]^-$ and HS$^-$ thermodynamic data were incorrect and that the major species controlling both sulfur and gold transport in hydrothermal-magmatic fluids is the HS$^-$ anion. Here I demonstrate that their conclusions stem from the use of inconsistent thermodynamic data and incorrect Raman spectra assignments. Thus, ref. 1 provides no grounds for questioning the validity of the thermodynamic properties of HS$^-$ and $[S_3^\bullet]^-$ or of their metal complexes in the fluids of the Earth's crust. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:05:15 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Pokrovski",
"Gleb S.",
"",
"GET"
]
] |
2506.08599 | Nicol\'as Hinrichs | Nicolas Hinrichs, Mahault Albarracin, Dimitris Bolis, Yuyue Jiang, Leonardo Christov-Moore, Leonhard Schilbach | Geometric Hyperscanning under Active Inference | 12 pages excl. references, 2 figures, submitted to the 6th International Workshop on Active Inference | null | null | null | q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Second-person neuroscience holds social cognition as embodied meaning co-regulation through reciprocal interaction, modeled here as coupled active inference with affect emerging as inference over identity-relevant surprise. Each agent maintains a self-model that tracks violations in its predictive coherence while recursively modeling the other. Valence is computed from self-model prediction error, weighted by self-relevance, and modulated by prior affective states and by what we term temporal aiming, which captures affective appraisal over time. This accommodates shifts in the self-other boundary, allowing affect to emerge at individual and dyadic levels. We propose a novel method termed geometric hyperscanning, based on the Forman-Ricci curvature, to empirically operationalize these processes: it tracks topological reconfigurations in inter-brain networks, with its entro-py serving as a proxy for affective phase transitions such as rupture, co-regulation, and re-attunement. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:06:28 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Hinrichs",
"Nicolas",
""
],
[
"Albarracin",
"Mahault",
""
],
[
"Bolis",
"Dimitris",
""
],
[
"Jiang",
"Yuyue",
""
],
[
"Christov-Moore",
"Leonardo",
""
],
[
"Schilbach",
"Leonhard",
""
]
] |
2506.08600 | Hiroshi Kera | Hiroshi Kera, Shun Arakawa, Yuta Sato | CALT: A Library for Computer Algebra with Transformer | ISSAC 2025 Short Communications | null | null | null | cs.LG cs.SC math.AC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in artificial intelligence have demonstrated the learnability of symbolic computation through end-to-end deep learning. Given a sufficient number of examples of symbolic expressions before and after the target computation, Transformer models - highly effective learners of sequence-to-sequence functions - can be trained to emulate the computation. This development opens up several intriguing challenges and new research directions, which require active contributions from the symbolic computation community. In this work, we introduce Computer Algebra with Transformer (CALT), a user-friendly Python library designed to help non-experts in deep learning train models for symbolic computation tasks. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:06:41 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Kera",
"Hiroshi",
""
],
[
"Arakawa",
"Shun",
""
],
[
"Sato",
"Yuta",
""
]
] |
2506.08601 | Toki Ikeda | Toki Ikeda, Takashi Shimonishi, Natsuko Izumi, Hiroyuki Kaneko, Satoko Takahashi, Kei E. I. Tanaka, Kenji Furuya, and Chikako Yasui | The detection of spatially resolved protosteller outflows and episodic jets in the outer Galaxy | 19 pages, 8 figures, accepted for publication in The Astrophysical Journal | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | We present the first detection of spatially resolved protostellar outflows and jets in the outer Galaxy. We observed five star-forming regions in the outer Galaxy (Sh 2--283, NOMF05-16/19/23/63; galactocentric distance = 15.7--17.4 kpc) with the Atacama Large Millimeter/submillimeter Array (ALMA). Towards Sh 2--283, we have detected distinct outflow ($\sim$5--50 km s$^{-1}$) and jet components ($\sim$50--100 km s$^{-1}$) associated with the protostar in CO(3--2) emission. The outflows and jets are well-collimated, with the jets exhibiting multiple bullet structures. The position-velocity diagram along the CO flow axis shows two characteristic structures: (a) the flow velocity which linearly increases with the position offset from the core center (Hubble-like flow), and (b) continuous velocity components of the periodical flows (spine-like structures), which may indicate the episodic mass-ejection event. The time intervals of the mass-ejection events are estimated to be 900--4000 years based on the slopes of these spine-like structures. These characteristics align with those of nearby protostellar systems, indicating that early star formation in low-metallicity environments, such as the outer Galaxy, resembles that in the inner Galaxy. In contrast to the physical similarities, the $N\mathrm{(SiO)}$/$N\mathrm{(CO)}$ ratio in the jet bullet appears to be lower than that measured in the low-mass protostellar sources in the inner Galaxy. This may indicate the different shock chemistry or different dust composition in the outer Galaxy source, although non-LTE effects could also affect the observed low $N\mathrm{(SiO)}$/$N\mathrm{(CO)}$ ratio. We also report the new detection of the other 4 outflow sources in the outer Galaxy. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:06:43 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Ikeda",
"Toki",
""
],
[
"Shimonishi",
"Takashi",
""
],
[
"Izumi",
"Natsuko",
""
],
[
"Kaneko",
"Hiroyuki",
""
],
[
"Takahashi",
"Satoko",
""
],
[
"Tanaka",
"Kei E. I.",
""
],
[
"Furuya",
"Kenji",
""
],
[
"Yasui",
"Chikako",
""
]
] |
2506.08602 | Tingzhi Li | Tingzhi Li, Xuefeng Liu | WGLE:Backdoor-free and Multi-bit Black-box Watermarking for Graph Neural Networks | null | null | null | null | cs.CR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Neural Networks (GNNs) are increasingly deployed in graph-related applications, making ownership verification critical to protect their intellectual property against model theft. Fingerprinting and black-box watermarking are two main methods. However, the former relies on determining model similarity, which is computationally expensive and prone to ownership collisions after model post-processing such as model pruning or fine-tuning. The latter embeds backdoors, exposing watermarked models to the risk of backdoor attacks. Moreover, both methods enable ownership verification but do not convey additional information. As a result, each distributed model requires a unique trigger graph, and all trigger graphs must be used to query the suspect model during verification. Multiple queries increase the financial cost and the risk of detection.
To address these challenges, this paper proposes WGLE, a novel black-box watermarking paradigm for GNNs that enables embedding the multi-bit string as the ownership information without using backdoors. WGLE builds on a key insight we term Layer-wise Distance Difference on an Edge (LDDE), which quantifies the difference between the feature distance and the prediction distance of two connected nodes. By predefining positive or negative LDDE values for multiple selected edges, WGLE embeds the watermark encoding the intended information without introducing incorrect mappings that compromise the primary task. WGLE is evaluated on six public datasets and six mainstream GNN architectures along with state-of-the-art methods. The results show that WGLE achieves 100% ownership verification accuracy, an average fidelity degradation of 0.85%, comparable robustness against potential attacks, and low embedding overhead. The code is available in the repository. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:12:00 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Li",
"Tingzhi",
""
],
[
"Liu",
"Xuefeng",
""
]
] |
2506.08603 | Yves Aubry | Yves Aubry (IMATH, I2M, UTLN), Fabien Herbaut (UCA INSPE, UniCA, IMATH), Julien Monaldi (IMATH) | Maximal curves with respect to quadratic extensions over finite fields | null | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a detailed study of a canonical bound which relates the numbers of rational points of a curve over a finite field with that over its quadratic extension. Alternative proofs which relate to the variance enable to complete the inequality in a symmetrical way and to obtain optimal refinements. We focus on the curves reaching the bound, which we call Diophantinemaximal curves. We provide different characterizations and stress natural links with the curves which attain the Ihara bound. As consequences, we establish the list of such curves with low genus and we outline a maximality result which involves the Suzuki curves. At last we determine which polynomials correspond to the Jacobian of a Diophantine-maximal curve of genus 2. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:12:07 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Aubry",
"Yves",
"",
"IMATH, I2M, UTLN"
],
[
"Herbaut",
"Fabien",
"",
"UCA INSPE, UniCA, IMATH"
],
[
"Monaldi",
"Julien",
"",
"IMATH"
]
] |
2506.08604 | Giacomo Baldan | Giacomo Baldan, Qiang Liu, Alberto Guardone, Nils Thuerey | Flow Matching Meets PDEs: A Unified Framework for Physics-Constrained Generation | null | null | null | null | cs.LG cs.AI cs.CE cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | Generative machine learning methods, such as diffusion models and flow matching, have shown great potential in modeling complex system behaviors and building efficient surrogate models. However, these methods typically learn the underlying physics implicitly from data. We propose Physics-Based Flow Matching (PBFM), a novel generative framework that explicitly embeds physical constraints, both PDE residuals and algebraic relations, into the flow matching objective. We also introduce temporal unrolling at training time that improves the accuracy of the final, noise-free sample prediction. Our method jointly minimizes the flow matching loss and the physics-based residual loss without requiring hyperparameter tuning of their relative weights. Additionally, we analyze the role of the minimum noise level, $\sigma_{\min}$, in the context of physical constraints and evaluate a stochastic sampling strategy that helps to reduce physical residuals. Through extensive benchmarks on three representative PDE problems, we show that our approach yields up to an $8\times$ more accurate physical residuals compared to FM, while clearly outperforming existing algorithms in terms of distributional accuracy. PBFM thus provides a principled and efficient framework for surrogate modeling, uncertainty quantification, and accelerated simulation in physics and engineering applications. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:13:37 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Baldan",
"Giacomo",
""
],
[
"Liu",
"Qiang",
""
],
[
"Guardone",
"Alberto",
""
],
[
"Thuerey",
"Nils",
""
]
] |
2506.08605 | Elena V. Mikheeva | A.N. Melikhov (P.N. Lebedev Physical Institute, Moscow, Russia), E.V. Mikheeva (Astro Space Center of the P.N. Lebedev Physical Institute, Moscow, Russia) | Constraints on primordial black holes from interstellar dust temperature in the Galaxy | 5 pages, 2 figures; typos in the bibliography are fixed | Physical Review D 107, 063535 (2023) | 10.1103/PhysRevD.107.063535 | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The photons emitted by primordial black holes (PBHs) can be a heating factor for interstellar dust. Assuming that dust in the Galaxy has a homogeneous distribution and that PBHs have the same distribution as dark matter (DM), we constrain the fraction of PBHs in DM from the observational dust temperature in the Galaxy. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:13:56 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Melikhov",
"A. N.",
"",
"P.N. Lebedev Physical Institute, Moscow, Russia"
],
[
"Mikheeva",
"E. V.",
"",
"Astro Space Center of the P.N. Lebedev Physical Institute, Moscow, Russia"
]
] |
2506.08606 | Radoslaw Klimek | Radoslaw Klimek | RE-oriented Model Development with LLM Support and Deduction-based Verification | The paper has been peer-reviewed and accepted for publication to the 1st International Workshop on Artificial Intelligence for Integrated Development Environments (AI-IDE) of the 33rd ACM Symposium on the Foundations of Software Engineering (FSE '25), June 23--27, 2025, Trondheim, Norway | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | The requirements engineering (RE) phase is pivotal in developing high-quality software. Integrating advanced modelling techniques with large language models (LLMs) and formal verification in a logical style can significantly enhance this process. We propose a comprehensive framework that focuses on specific Unified Modelling Language (UML) diagrams for preliminary system development. This framework offers visualisations at various modelling stages and seamlessly integrates large language models and logical reasoning engines. The behavioural models generated with the assistance of LLMs are automatically translated into formal logical specifications. Deductive formal verification ensures that logical requirements and interrelations between software artefacts are thoroughly addressed. Ultimately, the framework facilitates the automatic generation of program skeletons, streamlining the transition from design to implementation. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:13:58 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Klimek",
"Radoslaw",
""
]
] |
2506.08607 | Venktesh V | Kiran Purohit, V Venktesh, Sourangshu Bhattacharya and Avishek Anand | Sample Efficient Demonstration Selection for In-Context Learning | Accepted at ICML 2025 , 24 pages | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The in-context learning paradigm with LLMs has been instrumental in advancing a wide range of natural language processing tasks. The selection of few-shot examples (exemplars / demonstration samples) is essential for constructing effective prompts under context-length budget constraints. In this paper, we formulate the exemplar selection task as a top-m best arms identification problem. A key challenge in this setup is the exponentially large number of arms that need to be evaluated to identify the m-best arms. We propose CASE (Challenger Arm Sampling for Exemplar selection), a novel sample-efficient selective exploration strategy that maintains a shortlist of "challenger" arms, which are current candidates for the top-m arms. In each iteration, only one of the arms from this shortlist or the current topm set is pulled, thereby reducing sample complexity and, consequently, the number of LLM evaluations. Furthermore, we model the scores of exemplar subsets (arms) using a parameterized linear scoring function, leading to stochastic linear bandits setting. CASE achieves remarkable efficiency gains of up to 7x speedup in runtime while requiring 7x fewer LLM calls (87% reduction) without sacrificing performance compared to state-of-the-art exemplar selection methods. We release our code and data at https://github.com/kiranpurohit/CASE | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:16:11 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Purohit",
"Kiran",
""
],
[
"Venktesh",
"V",
""
],
[
"Bhattacharya",
"Sourangshu",
""
],
[
"Anand",
"Avishek",
""
]
] |
2506.08608 | Lv Yang | Yang Lv, Rong Hu, Bin Qian, Jian-Bo Yang | Q-learning-based Hierarchical Cooperative Local Search for Steelmaking-continuous Casting Scheduling Problem | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | The steelmaking continuous casting scheduling problem (SCCSP) is a critical and complex challenge in modern steel production, requiring the coordinated assignment and sequencing of steel charges across multiple production stages. Efficient scheduling not only enhances productivity but also significantly reduces energy consumption. However, both traditional heuristics (e.g., two-stage local search) and recent metaheuristics often struggle to adapt to the dynamic characteristics of practical SCCSP instances. To address these limitations, this paper introduces a novel Q learning based hierarchical cooperative local search framework, termed HierC_Q, aimed at minimizing the weighted sum of the maximum completion time and the average waiting time in SCCSP. The core contributions of HierC_Q are twofold. First, considering the intrinsic coupling properties of the SCCSP, a dedicated reward function is proposed based on a novel coupling measure (CM), guiding the exploration process towards promising regions of the solution space. Second, a hierarchical architecture is devised, comprising two distinct tiers: the learn to improve (L2I) tier and the "disturb to renovate" (D2R) tier. The L2I tier performs deep exploitation within promising regions using two independent Q-learning-based local search frameworks (QLSFs) tailored for subproblems, along with a synergy QLSF designed for the main problem. To enhance the effectiveness of local search, a validity evaluation approach and a speed-up evaluation method are also intro-duced, grounded in a detailed study of the problem's structure. Meanwhile, the D2R tier incorporates a perturbation and construction based solution renewal strategy to mitigate the risk of premature convergence. The superiority and effectiveness of HierC_Q are demonstrated through extensive comparisons with eleven local search frameworks and nine state-of-the-art algorithms. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:16:33 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Lv",
"Yang",
""
],
[
"Hu",
"Rong",
""
],
[
"Qian",
"Bin",
""
],
[
"Yang",
"Jian-Bo",
""
]
] |
2506.08609 | Xiaoning Feng | Xiaoning Feng, Hans Hon Sang Chan, David P. Tew | Quantum Resource Assay for the Grid-Based Simulation of the Photodynamics of Pyrazine | 20 pages, 24 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We establish and analyse the performance and resource requirements of an end-to-end fault-tolerant quantum algorithm for computing the absorption spectrum and population dynamics of photoexcited pyrazine. The quantum circuit construction consists of initial state preparation using uniformly controlled rotations, the time-dependent Hamiltonian propagation based on the grid-based Split Operator Quantum Fourier Transform (SO-QFT) method, and cost-effective measurements including statistical and canonical phase estimation. We use classical emulations to validate the quantum resources required for the task, and propose generalised formulae for the qubit count and gate depth calculation. Simulating the vibronic dynamics of pyrazine in a low-dimensional abstraction requires 17-qubit circuits with a gate depth of $\mathcal{O}(10^4)$, whereas a full-dimensional simulation of pyrazine in 24 modes requires at least 97-qubit circuits with a gate depth of $\mathcal{O}(10^5)$. Our work provides a foundational framework for understanding high-dimensional wavepacket-based quantum simulations of photo-induced dynamics and vibronic spectra, anticipating future applications in the simulation of even larger molecular systems on fault-tolerant quantum computers. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:17:02 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Feng",
"Xiaoning",
""
],
[
"Chan",
"Hans Hon Sang",
""
],
[
"Tew",
"David P.",
""
]
] |
2506.08610 | Federica Giacchino | Federica Giacchino | Axion-like particle constraints from preSN in future experiments | Proceeding of 2nd General Meeting of COST Action COSMIC WISPers (CA21106) | PoS (2024) volume 474, pages 032 | 10.22323/1.474.0032 | null | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We extend the study of the search for ALP-photon emissions from a Supernovae (SNe) progenitor, the Red Supergiant Star (RGS), as Betelgeuse, arising through a combination of Bremsstrahlung, Compton, and Primako? processes, in the hard-X and MeV energy range, using the next proposed and future experiments, such as COMCUBE, GECCO, COSI, AMEGO-X, and e-ASTROGAM. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:19:49 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Giacchino",
"Federica",
""
]
] |
2506.08611 | Shiji Zhao | Shiji Zhao, Chi Chen, Ranjie Duan, Xizhe Wang, Xingxing Wei | Towards Class-wise Fair Adversarial Training via Anti-Bias Soft Label Distillation | arXiv admin note: text overlap with arXiv:2312.05508 | null | null | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | Adversarial Training (AT) is widely recognized as an effective approach to enhance the adversarial robustness of Deep Neural Networks. As a variant of AT, Adversarial Robustness Distillation (ARD) has shown outstanding performance in enhancing the robustness of small models. However, both AT and ARD face robust fairness issue: these models tend to display strong adversarial robustness against some classes (easy classes) while demonstrating weak adversarial robustness against others (hard classes). This paper explores the underlying factors of this problem and points out the smoothness degree of soft labels for different classes significantly impacts the robust fairness from both empirical observation and theoretical analysis. Based on the above exploration, we propose Anti-Bias Soft Label Distillation (ABSLD) within the Knowledge Distillation framework to enhance the adversarial robust fairness. Specifically, ABSLD adaptively reduces the student's error risk gap between different classes, which is accomplished by adjusting the class-wise smoothness degree of teacher's soft labels during the training process, and the adjustment is managed by assigning varying temperatures to different classes. Additionally, as a label-based approach, ABSLD is highly adaptable and can be integrated with the sample-based methods. Extensive experiments demonstrate ABSLD outperforms state-of-the-art methods on the comprehensive performance of robustness and fairness. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:20:34 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Zhao",
"Shiji",
""
],
[
"Chen",
"Chi",
""
],
[
"Duan",
"Ranjie",
""
],
[
"Wang",
"Xizhe",
""
],
[
"Wei",
"Xingxing",
""
]
] |
2506.08612 | Robert-Jan Bruintjes | Robert-Jan Bruintjes, Attila Lengyel, Osman Semih Kayhan, Davide Zambrano, Nergis T\"omen, Hadi Jamali-Rad, Jan van Gemert | Data-Efficient Challenges in Visual Inductive Priors: A Retrospective | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Deep Learning requires large amounts of data to train models that work well. In data-deficient settings, performance can be degraded. We investigate which Deep Learning methods benefit training models in a data-deficient setting, by organizing the "VIPriors: Visual Inductive Priors for Data-Efficient Deep Learning" workshop series, featuring four editions of data-impaired challenges. These challenges address the problem of training deep learning models for computer vision tasks with limited data. Participants are limited to training models from scratch using a low number of training samples and are not allowed to use any form of transfer learning. We aim to stimulate the development of novel approaches that incorporate prior knowledge to improve the data efficiency of deep learning models. Successful challenge entries make use of large model ensembles that mix Transformers and CNNs, as well as heavy data augmentation. Novel prior knowledge-based methods contribute to success in some entries. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:21:48 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Bruintjes",
"Robert-Jan",
""
],
[
"Lengyel",
"Attila",
""
],
[
"Kayhan",
"Osman Semih",
""
],
[
"Zambrano",
"Davide",
""
],
[
"Tömen",
"Nergis",
""
],
[
"Jamali-Rad",
"Hadi",
""
],
[
"van Gemert",
"Jan",
""
]
] |
2506.08613 | Joost Van Dalen | Joost van Dalen, Yuki M. Asano, Marc Russwurm | SAMSelect: A Spectral Index Search for Marine Debris Visualization using Segment Anything | null | null | 10.1109/LGRS.2025.3572407 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This work proposes SAMSelect, an algorithm to obtain a salient three-channel visualization for multispectral images. We develop SAMSelect and show its use for marine scientists visually interpreting floating marine debris in Sentinel-2 imagery. These debris are notoriously difficult to visualize due to their compositional heterogeneity in medium-resolution imagery. Out of these difficulties, a visual interpretation of imagery showing marine debris remains a common practice by domain experts, who select bands and spectral indices on a case-by-case basis informed by common practices and heuristics. SAMSelect selects the band or index combination that achieves the best classification accuracy on a small annotated dataset through the Segment Anything Model. Its central assumption is that the three-channel visualization achieves the most accurate segmentation results also provide good visual information for photo-interpretation.
We evaluate SAMSelect in three Sentinel-2 scenes containing generic marine debris in Accra, Ghana, and Durban, South Africa, and deployed plastic targets from the Plastic Litter Project. This reveals the potential of new previously unused band combinations (e.g., a normalized difference index of B8, B2), which demonstrate improved performance compared to literature-based indices. We describe the algorithm in this paper and provide an open-source code repository that will be helpful for domain scientists doing visual photo interpretation, especially in the marine field. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:22:01 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"van Dalen",
"Joost",
""
],
[
"Asano",
"Yuki M.",
""
],
[
"Russwurm",
"Marc",
""
]
] |
2506.08614 | Mareike Fischer | Mareike Fischer and Tom Niklas Hamann and Kristina Wicke | Metaconcepts of rooted tree balance | null | null | null | null | q-bio.PE math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Measures of tree balance play an important role in many different research areas such as mathematical phylogenetics or theoretical computer science. Typically, tree balance is quantified by a single number which is assigned to the tree by a balance or imbalance index, of which several exist in the literature. Most of these indices are based on structural aspects of tree shape, such as clade sizes or leaf depths. For instance, indices like the Sackin index, total cophenetic index, and $\widehat{s}$-shape statistic all quantify tree balance through clade sizes, albeit with different definitions and properties.
In this paper, we formalize the idea that many tree (im)balance indices are functions of similar underlying tree shape characteristics by introducing metaconcepts of tree balance. A metaconcept is a function $\Phi_f$ that depends on a function $f$ capturing some aspect of tree shape, such as balance values, clade sizes, or leaf depths. These metaconcepts encompass existing indices but also provide new means of measuring tree balance. The versatility and generality of metaconcepts allow for the systematic study of entire families of (im)balance indices, providing deeper insights that extend beyond index-by-index analysis. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:23:10 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Fischer",
"Mareike",
""
],
[
"Hamann",
"Tom Niklas",
""
],
[
"Wicke",
"Kristina",
""
]
] |
2506.08615 | Jintian Lin | Xiaochao Luo, Chuntao Li, Xingzhao Huang, Jintian Lin, Renhong Gao, Yifei Yao, Yingnuo Qiu, Yixuan Yang, Lei Wang, Huakang Yu, Ya Cheng | Visible Brillouin-quadratic microlaser in a high-Q thin-film lithium niobate microdisk | 13 pages and 6 figures. arXiv admin note: substantial text overlap with arXiv:2503.23706 | null | null | null | physics.optics | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Narrow-linewidth lasers at short/visible wavelengths are crucial for quantum and atomic applications, such as atomic clocks, quantum computing, atomic and molecular spectroscopy, and quantum sensing. However, such lasers are often only accessible in bulky tabletop systems and remain scarce in integrated photonic platform. Here, we report an on-chip visible Brillouin-quadratic microlaser in a 117-um-diameter thin-film lithium niobate (TFLN) microdisk via dispersion engineering. Enabled by the ultra-high Q factor of 4.0X10(6) and small mode volume, strong photon-phonon interaction and high second-order nonlinearity of the TFLN microdisk, narrow-linewidth Stokes Brillouin lasing (SBL) is demonstrated with 10.17 GHz Brillouin shift under a 1560-nm pump, exhibiting a short-term narrow linewidth of 254 Hz and a low threshold of only 1.81 mW. Meanwhile, efficient second harmonic generation (SHG) of the SBL signal is also observed at 780 nm, with a normalized conversion efficiency of 3.61%/mW, made possible by simultaneous phase matching fulfillments for both narrow-linewidth SBL and its SHG. This demonstration of an integrated ultra-narrow linewidth visible wavelength Brillouin-quadratic lasers opens new avenues toward chip-scale quantum information processing and precise metrology. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:23:14 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Luo",
"Xiaochao",
""
],
[
"Li",
"Chuntao",
""
],
[
"Huang",
"Xingzhao",
""
],
[
"Lin",
"Jintian",
""
],
[
"Gao",
"Renhong",
""
],
[
"Yao",
"Yifei",
""
],
[
"Qiu",
"Yingnuo",
""
],
[
"Yang",
"Yixuan",
""
],
[
"Wang",
"Lei",
""
],
[
"Yu",
"Huakang",
""
],
[
"Cheng",
"Ya",
""
]
] |
2506.08616 | Gilles Bareilles | Julien Fageot, Peva Blanchard, Gilles Bareilles, L\^e-Nguy\^en Hoang | Generalizing while preserving monotonicity in comparison-based preference learning models | null | null | null | null | math.ST cs.LG stat.TH | http://creativecommons.org/licenses/by/4.0/ | If you tell a learning model that you prefer an alternative $a$ over another alternative $b$, then you probably expect the model to be monotone, that is, the valuation of $a$ increases, and that of $b$ decreases. Yet, perhaps surprisingly, many widely deployed comparison-based preference learning models, including large language models, fail to have this guarantee. Until now, the only comparison-based preference learning algorithms that were proved to be monotone are the Generalized Bradley-Terry models. Yet, these models are unable to generalize to uncompared data. In this paper, we advance the understanding of the set of models with generalization ability that are monotone. Namely, we propose a new class of Linear Generalized Bradley-Terry models with Diffusion Priors, and identify sufficient conditions on alternatives' embeddings that guarantee monotonicity. Our experiments show that this monotonicity is far from being a general guarantee, and that our new class of generalizing models improves accuracy, especially when the dataset is limited. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:24:08 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Fageot",
"Julien",
""
],
[
"Blanchard",
"Peva",
""
],
[
"Bareilles",
"Gilles",
""
],
[
"Hoang",
"Lê-Nguyên",
""
]
] |
2506.08617 | Naoto Yokoi | Naoto Yokoi, Yuki Tanaka, Yukito Nonaka, Shunsuke Daimon, Junji Haruyama, and Eiji Saitoh | Diffusion model for analyzing quantum fingerprints in conductance fluctuation | 6 pages, 5 figures | null | null | null | cond-mat.mes-hall quant-ph | http://creativecommons.org/licenses/by/4.0/ | A conditional diffusion model has been developed to analyze intricate conductance fluctuations called universal conductance fluctuations or quantum fingerprints appearing in quantum transport phenomena. The model reconstructs impurity arrangements and quantum interference patterns in nanometals by using magnetoconductance data, providing a novel approach to analyze complex data based on machine learning. In addition, we visualize the attention weights in the model, which efficiently extract information on the non-local correlation of the electron wave functions, and the score functions, which represent the force fields in the wave-function space. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:25:06 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Yokoi",
"Naoto",
""
],
[
"Tanaka",
"Yuki",
""
],
[
"Nonaka",
"Yukito",
""
],
[
"Daimon",
"Shunsuke",
""
],
[
"Haruyama",
"Junji",
""
],
[
"Saitoh",
"Eiji",
""
]
] |
2506.08618 | Xianquan Yan | Xianquan Yan, Hakan Akg\"un, Kenji Kawaguchi, N. Duane Loh, Ching Hua Lee | HSG-12M: A Large-Scale Spatial Multigraph Dataset | 39 pages, 13 figures, 3 tables. Code & pipeline: [https://github.com/sarinstein-yan/Poly2Graph] Dataset: [https://github.com/sarinstein-yan/HSG-12M] Dataset released under CC BY 4.0 | null | null | null | cs.LG cond-mat.mes-hall cond-mat.other cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing graph benchmarks assume non-spatial, simple edges, collapsing physically distinct paths into a single link. We introduce HSG-12M, the first large-scale dataset of $\textbf{spatial multigraphs}-$graphs embedded in a metric space where multiple geometrically distinct trajectories between two nodes are retained as separate edges. HSG-12M contains 11.6 million static and 5.1 million dynamic $\textit{Hamiltonian spectral graphs}$ across 1401 characteristic-polynomial classes, derived from 177 TB of spectral potential data. Each graph encodes the full geometry of a 1-D crystal's energy spectrum on the complex plane, producing diverse, physics-grounded topologies that transcend conventional node-coordinate datasets. To enable future extensions, we release $\texttt{Poly2Graph}$: a high-performance, open-source pipeline that maps arbitrary 1-D crystal Hamiltonians to spectral graphs. Benchmarks with popular GNNs expose new challenges in learning from multi-edge geometry at scale. Beyond its practical utility, we show that spectral graphs serve as universal topological fingerprints of polynomials, vectors, and matrices, forging a new algebra-to-graph link. HSG-12M lays the groundwork for geometry-aware graph learning and new opportunities of data-driven scientific discovery in condensed matter physics and beyond. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:25:19 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Yan",
"Xianquan",
""
],
[
"Akgün",
"Hakan",
""
],
[
"Kawaguchi",
"Kenji",
""
],
[
"Loh",
"N. Duane",
""
],
[
"Lee",
"Ching Hua",
""
]
] |
2506.08619 | Gon\c{c}alo Dias Pais | Gon\c{c}alo Dias Pais and Valter Piedade and Moitreya Chatterjee and Marcus Greiff and Pedro Miraldo | A Probability-guided Sampler for Neural Implicit Surface Rendering | Accepted in ECCV 2024 | European Conference on Computer Vision 2024 (pp. 164-182) | 10.1007/978-3-031-72913-3_10 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Several variants of Neural Radiance Fields (NeRFs) have significantly improved the accuracy of synthesized images and surface reconstruction of 3D scenes/objects. In all of these methods, a key characteristic is that none can train the neural network with every possible input data, specifically, every pixel and potential 3D point along the projection rays due to scalability issues. While vanilla NeRFs uniformly sample both the image pixels and 3D points along the projection rays, some variants focus only on guiding the sampling of the 3D points along the projection rays. In this paper, we leverage the implicit surface representation of the foreground scene and model a probability density function in a 3D image projection space to achieve a more targeted sampling of the rays toward regions of interest, resulting in improved rendering. Additionally, a new surface reconstruction loss is proposed for improved performance. This new loss fully explores the proposed 3D image projection space model and incorporates near-to-surface and empty space components. By integrating our novel sampling strategy and novel loss into current state-of-the-art neural implicit surface renderers, we achieve more accurate and detailed 3D reconstructions and improved image rendering, especially for the regions of interest in any given scene. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:28:16 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Pais",
"Gonçalo Dias",
""
],
[
"Piedade",
"Valter",
""
],
[
"Chatterjee",
"Moitreya",
""
],
[
"Greiff",
"Marcus",
""
],
[
"Miraldo",
"Pedro",
""
]
] |
2506.08620 | Xinhua Mao Dr. | Xinhua Mao, Manyi Tao, Jixia Fan | Extended Spherical Geometry Algorithm for Spaceborne SAR Processing in Stripmap and TOPS Imaging Modes | 11 pages, 11 figures | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Spherical Geometry Algorithm (SGA) demonstrates superior capability in achieving efficient and precise spaceborne SAR image formation processing, even under challenging imaging conditions including non-linear radar trajectories and spherical Earth surface geometry. Nevertheless, the original SGA is specifically developed for spotlight SAR data processing and can't directly applied to processing spaceborne SAR data in other modes. In this paper, we first analyze the limitations of the SGA algorithm when applied to stripmap or TOPS mode SAR processing, and then propose an improved SGA algorithm which can process both stripmap and TOPS SAR data. Compared with the original algorithm, the new algorithm has two main differences. Firstly, in order to avoid undersampling during azimuth resampling in both modes, an instantaneous Doppler centroid removal process was added before azimuth interpolation processing by exploiting the endomorphism property of resampling operation. Secondly, the spectral analysis method used for the final step of azimuth compression in the original SGA has been replaced with a new matched filtering processing, which can avoid image aliasing in azimuth direction and improve computational efficiency. Measured real data processing results are presented to demonstrate the validity of the proposed algorithms. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:28:51 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Mao",
"Xinhua",
""
],
[
"Tao",
"Manyi",
""
],
[
"Fan",
"Jixia",
""
]
] |
2506.08621 | E. Bulut Kul | E. Bulut Kul, G\"okhan \"Oztarhan, M. N. \c{C}{\i}nar, A. D. G\"u\c{c}l\"u | Quantum Monte Carlo study of artificial triangular graphene quantum dots | 6 pages, 5 figures | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the magnetic phases of triangular graphene quantum
dots (TGQDs) with zigzag edges using variational and quantum Monte
Carlo methods. These systems serve as quantum simulators for
bipartite lattices with broken sublattice symmetry, providing a
platform to study the extended Hubbard model's emergent magnetic
phenomena, including Lieb's magnetism at half-filling, edge
depolarization upon single-electron addition, and Nagaoka
ferromagnetism. Our non-perturbative quantum Monte Carlo simulations,
performed for lattices of up to 61 sites, reveal that TGQDs
transition from metallic to insulating regimes as a function of site
radius size, while retaining edge-polarized ground states at
half-filling. Notably, edge depolarization occurs upon
single-electron doping in both metallic and insulating phases,
contrasting with the Nagaoka ferromagnetism observed in hexagonal
armchair geometries. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:31:25 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Kul",
"E. Bulut",
""
],
[
"Öztarhan",
"Gökhan",
""
],
[
"Çınar",
"M. N.",
""
],
[
"Güçlü",
"A. D.",
""
]
] |
2506.08623 | Rinat Prochii | Rinat Prochii, Elizaveta Dakhova, Pavel Birulin and Maxim Sharaev | Biologically Inspired Deep Learning Approaches for Fetal Ultrasound Image Classification | 16 pages, 2 figures, 3 tables | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate classification of second-trimester fetal ultrasound images remains challenging due to low image quality, high intra-class variability, and significant class imbalance. In this work, we introduce a simple yet powerful, biologically inspired deep learning ensemble framework that-unlike prior studies focused on only a handful of anatomical targets-simultaneously distinguishes 16 fetal structures. Drawing on the hierarchical, modular organization of biological vision systems, our model stacks two complementary branches (a "shallow" path for coarse, low-resolution cues and a "detailed" path for fine, high-resolution features), concatenating their outputs for final prediction. To our knowledge, no existing method has addressed such a large number of classes with a comparably lightweight architecture. We trained and evaluated on 5,298 routinely acquired clinical images (annotated by three experts and reconciled via Dawid-Skene), reflecting real-world noise and variability rather than a "cleaned" dataset. Despite this complexity, our ensemble (EfficientNet-B0 + EfficientNet-B6 with LDAM-Focal loss) identifies 90% of organs with accuracy > 0.75 and 75% of organs with accuracy > 0.85-performance competitive with more elaborate models applied to far fewer categories. These results demonstrate that biologically inspired modular stacking can yield robust, scalable fetal anatomy recognition in challenging clinical settings. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:34:33 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Prochii",
"Rinat",
""
],
[
"Dakhova",
"Elizaveta",
""
],
[
"Birulin",
"Pavel",
""
],
[
"Sharaev",
"Maxim",
""
]
] |
2506.08625 | Minhae Oh | Minhae Oh, Jeonghye Kim, Nakyung Lee, Donggeon Seo, Taeuk Kim, Jungwoo Lee | RAISE: Enhancing Scientific Reasoning in LLMs via Step-by-Step Retrieval | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Scientific reasoning requires not only long-chain reasoning processes, but also knowledge of domain-specific terminologies and adaptation to updated findings. To deal with these challenges for scientific reasoning, we introduce RAISE, a step-by-step retrieval-augmented framework which retrieves logically relevant documents from in-the-wild corpus. RAISE is divided into three steps: problem decomposition, logical query generation, and logical retrieval. We observe that RAISE consistently outperforms other baselines on scientific reasoning benchmarks. We analyze that unlike other baselines, RAISE retrieves documents that are not only similar in terms of the domain knowledge, but also documents logically more relevant. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:40:53 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Oh",
"Minhae",
""
],
[
"Kim",
"Jeonghye",
""
],
[
"Lee",
"Nakyung",
""
],
[
"Seo",
"Donggeon",
""
],
[
"Kim",
"Taeuk",
""
],
[
"Lee",
"Jungwoo",
""
]
] |
2506.08627 | Xixi Lu | Douwe Geurtjens, Xixi Lu | FoldA: Computing Partial-Order Alignments Using Directed Net Unfoldings | Conditionally accepted at BPM 2025 | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conformance checking is a fundamental task of process mining, which quantifies the extent to which the observed process executions match a normative process model. The state-of-the-art approaches compute alignments by exploring the state space formed by the synchronous product of the process model and the trace. This often leads to state space explosion, particularly when the model exhibits a high degree of choice and concurrency. Moreover, as alignments inherently impose a sequential structure, they fail to fully represent the concurrent behavior present in many real-world processes. To address these limitations, this paper proposes a new technique for computing partial-order alignments {on the fly using directed Petri net unfoldings, named FoldA. We evaluate our technique on 485 synthetic model-log pairs and compare it against Astar- and Dijkstra-alignments on 13 real-life model-log pairs and 6 benchmark pairs. The results show that our unfolding alignment, although it requires more computation time, generally reduces the number of queued states and provides a more accurate representation of concurrency. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:44:05 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Geurtjens",
"Douwe",
""
],
[
"Lu",
"Xixi",
""
]
] |
2506.08628 | Radoslaw Klimek | Radoslaw Klimek, Julia Witek | Logic Mining from Process Logs: Towards Automated Specification and Verification | This is a draft version of the article submitted to a journal from the JCR list | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Logical specifications play a key role in the formal analysis of behavioural models. Automating the derivation of such specifications is particularly valuable in complex systems, where manual construction is time-consuming and error-prone. This article presents an approach for generating logical specifications from process models discovered via workflow mining, combining pattern-based translation with automated reasoning techniques. In contrast to earlier work, we evaluate the method on both general-purpose and real-case event logs, enabling a broader empirical assessment. The study examines the impact of data quality, particularly noise, on the structure and testability of generated specifications. Using automated theorem provers, we validate a variety of logical properties, including satisfiability, internal consistency, and alignment with predefined requirements. The results support the applicability of the approach in realistic settings and its potential integration into empirical software engineering practices. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:44:19 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Klimek",
"Radoslaw",
""
],
[
"Witek",
"Julia",
""
]
] |
2506.08629 | Feixiang Du | Feixiang Du, Shengkun Wu | ECMNet:Lightweight Semantic Segmentation with Efficient CNN-Mamba Network | 16 pages, 2 figures, 4 tables | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the past decade, Convolutional Neural Networks (CNNs) and Transformers have achieved wide applicaiton in semantic segmentation tasks. Although CNNs with Transformer models greatly improve performance, the global context modeling remains inadequate. Recently, Mamba achieved great potential in vision tasks, showing its advantages in modeling long-range dependency. In this paper, we propose a lightweight Efficient CNN-Mamba Network for semantic segmentation, dubbed as ECMNet. ECMNet combines CNN with Mamba skillfully in a capsule-based framework to address their complementary weaknesses. Specifically, We design a Enhanced Dual-Attention Block (EDAB) for lightweight bottleneck. In order to improve the representations ability of feature, We devise a Multi-Scale Attention Unit (MSAU) to integrate multi-scale feature aggregation, spatial aggregation and channel aggregation. Moreover, a Mamba enhanced Feature Fusion Module (FFM) merges diverse level feature, significantly enhancing segmented accuracy. Extensive experiments on two representative datasets demonstrate that the proposed model excels in accuracy and efficiency balance, achieving 70.6% mIoU on Cityscapes and 73.6% mIoU on CamVid test datasets, with 0.87M parameters and 8.27G FLOPs on a single RTX 3090 GPU platform. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:44:23 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Du",
"Feixiang",
""
],
[
"Wu",
"Shengkun",
""
]
] |
2506.08630 | Laurens Engwegen | Laurens Engwegen, Daan Brinks, Wendelin B\"ohmer | Modular Recurrence in Contextual MDPs for Universal Morphology Control | null | null | null | null | cs.AI cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A universal controller for any robot morphology would greatly improve computational and data efficiency. By utilizing contextual information about the properties of individual robots and exploiting their modular structure in the architecture of deep reinforcement learning agents, steps have been made towards multi-robot control. Generalization to new, unseen robots, however, remains a challenge. In this paper we hypothesize that the relevant contextual information is partially observable, but that it can be inferred through interactions for better generalization to contexts that are not seen during training. To this extent, we implement a modular recurrent architecture and evaluate its generalization performance on a large set of MuJoCo robots. The results show a substantial improved performance on robots with unseen dynamics, kinematics, and topologies, in four different environments. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:44:30 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Engwegen",
"Laurens",
""
],
[
"Brinks",
"Daan",
""
],
[
"Böhmer",
"Wendelin",
""
]
] |
2506.08631 | Mohamed Camil Belhadjoudja | Mohamed Camil Belhadjoudja (GIPSA-INFINITY), M Maghenem (GIPSA-INFINITY), E Witrant (GIPSA-INFINITY), Didier Georges (GIPSA-INFINITY) | Boundary Control for Wildfire Mitigation | Submitted to the 2025 IEEE Conference on Decision and Control (CDC'25) | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a feedback control strategy to protect vulnerable areas from wildfires. We consider a system of coupled partial differential equations (PDEs) that models heat propagation and fuel depletion in wildfires and study two cases. First, when the wind velocity is known, we design a Neumann-type boundary controller guaranteeing that the temperature of some protected region converges exponentially, in the $L^2$ norm, to the ambient temperature. Second, when the wind velocity is unknown, we design an adaptive Neumann-type boundary controller guaranteeing the asymptotic convergence, in the $L^2$ norm, of the temperature of the protected region to the ambient temperature. In both cases, the controller acts along the boundary of the protected region and relies solely on temperature measurements along that boundary. Our results are supported by numerical simulations. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:45:30 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Belhadjoudja",
"Mohamed Camil",
"",
"GIPSA-INFINITY"
],
[
"Maghenem",
"M",
"",
"GIPSA-INFINITY"
],
[
"Witrant",
"E",
"",
"GIPSA-INFINITY"
],
[
"Georges",
"Didier",
"",
"GIPSA-INFINITY"
]
] |
2506.08632 | Yang Bai | Yang Bai, Liudi Yang, George Eskandar, Fengyi Shen, Dong Chen, Mohammad Altillawi, Ziyuan Liu, Gitta Kutyniok | RoboSwap: A GAN-driven Video Diffusion Framework For Unsupervised Robot Arm Swapping | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recent advancements in generative models have revolutionized video synthesis and editing. However, the scarcity of diverse, high-quality datasets continues to hinder video-conditioned robotic learning, limiting cross-platform generalization. In this work, we address the challenge of swapping a robotic arm in one video with another: a key step for crossembodiment learning. Unlike previous methods that depend on paired video demonstrations in the same environmental settings, our proposed framework, RoboSwap, operates on unpaired data from diverse environments, alleviating the data collection needs. RoboSwap introduces a novel video editing pipeline integrating both GANs and diffusion models, combining their isolated advantages. Specifically, we segment robotic arms from their backgrounds and train an unpaired GAN model to translate one robotic arm to another. The translated arm is blended with the original video background and refined with a diffusion model to enhance coherence, motion realism and object interaction. The GAN and diffusion stages are trained independently. Our experiments demonstrate that RoboSwap outperforms state-of-the-art video and image editing models on three benchmarks in terms of both structural coherence and motion consistency, thereby offering a robust solution for generating reliable, cross-embodiment data in robotic learning. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:46:07 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Bai",
"Yang",
""
],
[
"Yang",
"Liudi",
""
],
[
"Eskandar",
"George",
""
],
[
"Shen",
"Fengyi",
""
],
[
"Chen",
"Dong",
""
],
[
"Altillawi",
"Mohammad",
""
],
[
"Liu",
"Ziyuan",
""
],
[
"Kutyniok",
"Gitta",
""
]
] |
2506.08633 | \v{S}imon Sedl\'a\v{c}ek | \v{S}imon Sedl\'a\v{c}ek, Bolaji Yusuf, J\'an \v{S}vec, Pradyoth Hegde, Santosh Kesiraju, Old\v{r}ich Plchot, Jan \v{C}ernock\'y | Approaching Dialogue State Tracking via Aligning Speech Encoders and LLMs | Accepted to Interspeech 2025 | null | null | null | eess.AS cs.CL | http://creativecommons.org/licenses/by/4.0/ | In this work, we approach spoken Dialogue State Tracking (DST) by bridging the representation spaces of speech encoders and LLMs via a small connector module, with a focus on fully open-sourced and open-data components (WavLM-large, OLMo). We focus on ablating different aspects of such systems including full/LoRA adapter fine-tuning, the effect of agent turns in the dialogue history, as well as fuzzy matching-based output post-processing, which greatly improves performance of our systems on named entities in the dialogue slot values. We conduct our experiments on the SpokenWOZ dataset, and additionally utilize the Speech-Aware MultiWOZ dataset to augment our training data. Ultimately, our best-performing WavLM + connector + OLMo-1B aligned models achieve state of the art on the SpokenWOZ test set (34.66% JGA), and our system with Gemma-2-9B-instruct further surpasses this result, reaching 42.17% JGA on SpokenWOZ test. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:46:29 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Sedláček",
"Šimon",
""
],
[
"Yusuf",
"Bolaji",
""
],
[
"Švec",
"Ján",
""
],
[
"Hegde",
"Pradyoth",
""
],
[
"Kesiraju",
"Santosh",
""
],
[
"Plchot",
"Oldřich",
""
],
[
"Černocký",
"Jan",
""
]
] |
2506.08634 | Alvaro Becerra | Alvaro Becerra, Daniel Andres, Pablo Villegas, Roberto Daza, Ruth Cobos | MOSAIC-F: A Framework for Enhancing Students' Oral Presentation Skills through Personalized Feedback | Accepted in LASI Spain 25: Learning Analytics Summer Institute Spain 2025 | null | null | null | cs.HC cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this article, we present a novel multimodal feedback framework called MOSAIC-F, an acronym for a data-driven Framework that integrates Multimodal Learning Analytics (MMLA), Observations, Sensors, Artificial Intelligence (AI), and Collaborative assessments for generating personalized feedback on student learning activities. This framework consists of four key steps. First, peers and professors' assessments are conducted through standardized rubrics (that include both quantitative and qualitative evaluations). Second, multimodal data are collected during learning activities, including video recordings, audio capture, gaze tracking, physiological signals (heart rate, motion data), and behavioral interactions. Third, personalized feedback is generated using AI, synthesizing human-based evaluations and data-based multimodal insights such as posture, speech patterns, stress levels, and cognitive load, among others. Finally, students review their own performance through video recordings and engage in self-assessment and feedback visualization, comparing their own evaluations with peers and professors' assessments, class averages, and AI-generated recommendations. By combining human-based and data-based evaluation techniques, this framework enables more accurate, personalized and actionable feedback. We tested MOSAIC-F in the context of improving oral presentation skills. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:46:31 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Becerra",
"Alvaro",
""
],
[
"Andres",
"Daniel",
""
],
[
"Villegas",
"Pablo",
""
],
[
"Daza",
"Roberto",
""
],
[
"Cobos",
"Ruth",
""
]
] |
2506.08635 | Gon\c{c}alo Dias Pais | Siddhant Ranade and Gon\c{c}alo Dias Pais and Ross Tyler Whitaker and Jacinto C. Nascimento and Pedro Miraldo and Srikumar Ramalingam | SurfR: Surface Reconstruction with Multi-scale Attention | Accepted in 3DV 2025 | International Conference on 3D Vision 2025 | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We propose a fast and accurate surface reconstruction algorithm for unorganized point clouds using an implicit representation. Recent learning methods are either single-object representations with small neural models that allow for high surface details but require per-object training or generalized representations that require larger models and generalize to newer shapes but lack details, and inference is slow. We propose a new implicit representation for general 3D shapes that is faster than all the baselines at their optimum resolution, with only a marginal loss in performance compared to the state-of-the-art. We achieve the best accuracy-speed trade-off using three key contributions. Many implicit methods extract features from the point cloud to classify whether a query point is inside or outside the object. First, to speed up the reconstruction, we show that this feature extraction does not need to use the query point at an early stage (lazy query). Second, we use a parallel multi-scale grid representation to develop robust features for different noise levels and input resolutions. Finally, we show that attention across scales can provide improved reconstruction results. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:46:36 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Ranade",
"Siddhant",
""
],
[
"Pais",
"Gonçalo Dias",
""
],
[
"Whitaker",
"Ross Tyler",
""
],
[
"Nascimento",
"Jacinto C.",
""
],
[
"Miraldo",
"Pedro",
""
],
[
"Ramalingam",
"Srikumar",
""
]
] |
2506.08636 | Zhuolun Li | Zeinab Nezami, Zhuolun Li, Chuhao Qin, Fatemeh Banaie, Rabiya Khalid, Evangelos Pournaras | Blockchain and Edge Computing Nexus: A Large-scale Systematic Literature Review | null | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | Blockchain and edge computing are two instrumental paradigms of decentralized computation, driving key advancements in Smart Cities applications such as supply chain, energy and mobility. Despite their unprecedented impact on society, they remain significantly fragmented as technologies and research areas, while they share fundamental principles of distributed systems and domains of applicability. This paper introduces a novel and large-scale systematic literature review on the nexus of blockchain and edge computing with the aim to unravel a new understanding of how the interfacing of the two computing paradigms can boost innovation to provide solutions to timely but also long-standing research challenges. By collecting almost 6000 papers from 3 databases and putting under scrutiny almost 1000 papers, we build a novel taxonomy and classification consisting of 22 features with 287 attributes that we study using quantitative and machine learning methods. They cover a broad spectrum of technological, design, epistemological and sustainability aspects. Results reveal 4 distinguishing patterns of interplay between blockchain and edge computing with key determinants the public (permissionless) vs. private (permissioned) design, technology and proof of concepts. They also demonstrate the prevalence of blockchain-assisted edge computing for improving privacy and security, in particular for mobile computing applications. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:49:41 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Nezami",
"Zeinab",
""
],
[
"Li",
"Zhuolun",
""
],
[
"Qin",
"Chuhao",
""
],
[
"Banaie",
"Fatemeh",
""
],
[
"Khalid",
"Rabiya",
""
],
[
"Pournaras",
"Evangelos",
""
]
] |
2506.08637 | Bruno Torresani | Pierre Warion (I2M), Bruno Torr\'esani (I2M) | Stability estimates for adaptive focused time-frequency transforms | null | null | null | null | math.CA eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This contribution is a follow-up of a recent paper by the authors on adaptive, non-linear time-frequency transforms, focusing on the STFT based transforms. The adaptivity is provided by a focus function, that depends on the analyzed function or signal, and that adapts dynamically the time-frequency resolution of the analysis. Sticking to the continuous case setting, this work provides new stability results for the transform (stability with respect with the focus function). It also investigates in some details focus functions based upon regularized R{\'e}nyi entropies and show corresponding continuity results. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:50:03 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Warion",
"Pierre",
"",
"I2M"
],
[
"Torrésani",
"Bruno",
"",
"I2M"
]
] |
2506.08638 | Stian Backe | Amund Norland, Lasse Skare, Ole Jakob Viken, Stian Backe | Industrial Flexibility Investment Under Uncertainty: A Multi-Stage Stochastic Framework Considering Energy and Reserve Market Participation | 5 pages, 5 figures. This work has been submitted to the IEEE for possible publication (IEEE ISGT Europe 2025) | null | null | null | econ.GN q-fin.EC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The global energy transition toward net-zero emissions by 2050 is expected to increase the share of variable renewable energy sources (VRES) in the energy mix. As a result, industrial actors will encounter more complex market conditions, characterized by volatile electricity prices, rising carbon costs, and stricter regulations. This situation calls for the industry to capitalize on opportunities in both spot-price arbitrage and reserve market participation, while also meeting future regulatory demands. This paper presents a multi-stage optimization framework that supports investment decisions in flexible assets and enables reserve market participation by delivering ancillary services. The framework incorporates investment decisions, spot- and reserve-market bidding, and real-time operation. Uncertainty in market prices and operational conditions is handled through a nodal formulation. A case study of a large industrial site in Norway is performed, comparing the investment decisions with future technology- and carbon pricing scenarios under varying market conditions. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:52:24 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Norland",
"Amund",
""
],
[
"Skare",
"Lasse",
""
],
[
"Viken",
"Ole Jakob",
""
],
[
"Backe",
"Stian",
""
]
] |
2506.08639 | Amir Hossein Barjini | Amir Hossein Barjini, Seyed Adel Alizadeh Kolagar, Sadeq Yaqubi, Jouni Mattila | Deep Reinforcement Learning-Based Motion Planning and PDE Control for Flexible Manipulators | null | null | null | null | cs.RO cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | This article presents a motion planning and control framework for flexible robotic manipulators, integrating deep reinforcement learning (DRL) with a nonlinear partial differential equation (PDE) controller. Unlike conventional approaches that focus solely on control, we demonstrate that the desired trajectory significantly influences endpoint vibrations. To address this, a DRL motion planner, trained using the soft actor-critic (SAC) algorithm, generates optimized trajectories that inherently minimize vibrations. The PDE nonlinear controller then computes the required torques to track the planned trajectory while ensuring closed-loop stability using Lyapunov analysis. The proposed methodology is validated through both simulations and real-world experiments, demonstrating superior vibration suppression and tracking accuracy compared to traditional methods. The results underscore the potential of combining learning-based motion planning with model-based control for enhancing the precision and stability of flexible robotic manipulators. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:53:57 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Barjini",
"Amir Hossein",
""
],
[
"Kolagar",
"Seyed Adel Alizadeh",
""
],
[
"Yaqubi",
"Sadeq",
""
],
[
"Mattila",
"Jouni",
""
]
] |
2506.08640 | Yichong Lu | Yichong Lu, Yuzhuo Tian, Zijin Jiang, Yikun Zhao, Yuanbo Yang, Hao Ouyang, Haoji Hu, Huimin Yu, Yujun Shen, Yiyi Liao | Orientation Matters: Making 3D Generative Models Orientation-Aligned | Project Page: https://xdimlab.github.io/Orientation_Matters | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Humans intuitively perceive object shape and orientation from a single image, guided by strong priors about canonical poses. However, existing 3D generative models often produce misaligned results due to inconsistent training data, limiting their usability in downstream tasks. To address this gap, we introduce the task of orientation-aligned 3D object generation: producing 3D objects from single images with consistent orientations across categories. To facilitate this, we construct Objaverse-OA, a dataset of 14,832 orientation-aligned 3D models spanning 1,008 categories. Leveraging Objaverse-OA, we fine-tune two representative 3D generative models based on multi-view diffusion and 3D variational autoencoder frameworks to produce aligned objects that generalize well to unseen objects across various categories. Experimental results demonstrate the superiority of our method over post-hoc alignment approaches. Furthermore, we showcase downstream applications enabled by our aligned object generation, including zero-shot object orientation estimation via analysis-by-synthesis and efficient arrow-based object rotation manipulation. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:54:37 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Lu",
"Yichong",
""
],
[
"Tian",
"Yuzhuo",
""
],
[
"Jiang",
"Zijin",
""
],
[
"Zhao",
"Yikun",
""
],
[
"Yang",
"Yuanbo",
""
],
[
"Ouyang",
"Hao",
""
],
[
"Hu",
"Haoji",
""
],
[
"Yu",
"Huimin",
""
],
[
"Shen",
"Yujun",
""
],
[
"Liao",
"Yiyi",
""
]
] |
2506.08641 | Simon Roschmann | Simon Roschmann, Quentin Bouniot, Vasilii Feofanov, Ievgen Redko, Zeynep Akata | Time Series Representations for Classification Lie Hidden in Pretrained Vision Transformers | null | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Time series classification is a fundamental task in healthcare and industry, yet the development of time series foundation models (TSFMs) remains limited by the scarcity of publicly available time series datasets. In this work, we propose Time Vision Transformer (TiViT), a framework that converts time series into images to leverage the representational power of frozen Vision Transformers (ViTs) pretrained on large-scale image datasets. First, we theoretically motivate our approach by analyzing the 2D patching of ViTs for time series, showing that it can increase the number of label-relevant tokens and reduce the sample complexity. Second, we empirically demonstrate that TiViT achieves state-of-the-art performance on standard time series classification benchmarks by utilizing the hidden representations of large OpenCLIP models. We explore the structure of TiViT representations and find that intermediate layers with high intrinsic dimension are the most effective for time series classification. Finally, we assess the alignment between TiViT and TSFM representation spaces and identify a strong complementarity, with further performance gains achieved by combining their features. Our findings reveal yet another direction for reusing vision representations in a non-visual domain. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:54:51 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Roschmann",
"Simon",
""
],
[
"Bouniot",
"Quentin",
""
],
[
"Feofanov",
"Vasilii",
""
],
[
"Redko",
"Ievgen",
""
],
[
"Akata",
"Zeynep",
""
]
] |
2506.08642 | Italo Salas | Italo Salas, Francisca Guzm\'an-Lastra, Denisse Past\'en and Ariel Norambuena | Contact temporal network during motility induced phase separation | null | null | null | null | physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | Motility-induced phase separation (MIPS) is a paradigmatic non-equilibrium transition in active matter, determined by the P\'eclet number and packing fraction. We investigate the single-phase and phase-separated regimes of MIPS using a complex network approach, where networks are constructed from particle interactions over finite time windows. In the single-phase (gas-like) regime, the degree distributions $P(k)$ exhibit Gaussian behavior and resemble those of random graphs. Plotting the location and height of the $P(k)$ peak reveals a universal curve across different P\'eclet numbers at fixed packing fraction. In the phase-separated regime, we analyze the dense and dilute phases independently. The $P(k)$ distributions unveil distinct collective dynamics, including caging in the dense phase and the emergence of active solid-like structures at longer times. Clustering coefficients and average path lengths in both phases stabilize rapidly, indicating that short simulations are sufficient to capture essential network features. Overall, our results show that network metrics expose both universal and phase-specific aspects of active matter dynamics. Notably, we identify distinct and previously unreported topological structures arising in the dense and dilute phases within the MIPS regime. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:54:54 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Salas",
"Italo",
""
],
[
"Guzmán-Lastra",
"Francisca",
""
],
[
"Pastén",
"Denisse",
""
],
[
"Norambuena",
"Ariel",
""
]
] |
2506.08643 | Son The Nguyen | Son The Nguyen, Theja Tulabandhula | MEMETRON: Metaheuristic Mechanisms for Test-time Response Optimization of Large Language Models | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) are increasingly used for both open-ended and structured tasks, yet their inference-time behavior is still largely dictated by heuristic decoding strategies such as greedy search, sampling, or reranking. These methods provide limited control and do not explicitly optimize for task-specific objectives. We introduce MEMETRON, a task-agnostic framework that formulates LLM decoding as a discrete black-box optimization problem. MEMETRON leverages hybrid metaheuristic algorithms, GENETRON and ANNETRON, to search the response space, guided by reward models and contextual operations performed by the LLM itself. This approach enables efficient discovery of high-reward responses without requiring model retraining or gradient access. The framework is modular and generalizes across diverse tasks, requiring only a reward function and lightweight prompt templates. We evaluate our framework on the critical human preference alignment task and demonstrate that it significantly outperforms standard decoding and reranking methods, highlighting its potential to improve alignment without model retraining. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:55:53 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Nguyen",
"Son The",
""
],
[
"Tulabandhula",
"Theja",
""
]
] |
2506.08644 | Woosung Kim | Woosung Kim, JunHo Seo, Jongmin Lee, Byung-Jun Lee | Semi-gradient DICE for Offline Constrained Reinforcement Learning | Constrained Offline Reinforcement Learning | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Stationary Distribution Correction Estimation (DICE) addresses the mismatch between the stationary distribution induced by a policy and the target distribution required for reliable off-policy evaluation (OPE) and policy optimization. DICE-based offline constrained RL particularly benefits from the flexibility of DICE, as it simultaneously maximizes return while estimating costs in offline settings. However, we have observed that recent approaches designed to enhance the offline RL performance of the DICE framework inadvertently undermine its ability to perform OPE, making them unsuitable for constrained RL scenarios. In this paper, we identify the root cause of this limitation: their reliance on a semi-gradient optimization, which solves a fundamentally different optimization problem and results in failures in cost estimation. Building on these insights, we propose a novel method to enable OPE and constrained RL through semi-gradient DICE. Our method ensures accurate cost estimation and achieves state-of-the-art performance on the offline constrained RL benchmark, DSRL. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:57:25 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Kim",
"Woosung",
""
],
[
"Seo",
"JunHo",
""
],
[
"Lee",
"Jongmin",
""
],
[
"Lee",
"Byung-Jun",
""
]
] |
2506.08645 | Youqi Wu | Youqi Wu, Jingwei Zhang, Farzan Farnia | Fusing Cross-modal and Uni-modal Representations: A Kronecker Product Approach | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Cross-modal embeddings, such as CLIP, BLIP and their variants, have achieved promising results in aligning representations across modalities. However, these embeddings could underperform compared to state-of-the-art single-modality embeddings on modality-specific tasks. On the other hand, single-modality embeddings excel in their domains but lack cross-modal alignment capabilities. In this work, we focus on the problem of unifying cross-modality and single-modality embeddings to achieve the performance of modality-expert embedding within individual modalities while preserving cross-modal alignment. To this end, we propose RP-KrossFuse, a method that leverages a random projection-based Kronecker product to integrate cross-modal embeddings with single-modality embeddings. RP-KrossFuse aims to fuse the sample-pairwise similarity scores of the fused embeddings and operates efficiently in a specified kernel space and supports scalable implementations via random Fourier features for shift-invariant kernels such as the Gaussian kernel. We demonstrate the effectiveness of RP-KrossFuse through several numerical experiments, combining CLIP embeddings with uni-modal image and text embeddings. Our numerical results indicate that RP-KrossFuse achieves competitive modality-specific performance while retaining cross-modal alignment, bridging the gap between cross-modal and single-modality embeddings. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:57:58 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Wu",
"Youqi",
""
],
[
"Zhang",
"Jingwei",
""
],
[
"Farnia",
"Farzan",
""
]
] |
2506.08646 | Mingyu Zheng | Mingyu Zheng, Zhifan Feng, Jia Wang, Lanrui Wang, Zheng Lin, Yang Hao, Weiping Wang | TableDreamer: Progressive and Weakness-guided Data Synthesis from Scratch for Table Instruction Tuning | 27 pages, 19 figures, Findings of ACL 2025 | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Despite the commendable progress of recent LLM-based data synthesis methods, they face two limitations in generating table instruction tuning data. First, they can not thoroughly explore the vast input space of table understanding tasks, leading to limited data diversity. Second, they ignore the weaknesses in table understanding ability of the target LLM and blindly pursue the increase of data quantity, resulting in suboptimal data efficiency. In this paper, we introduce a progressive and weakness-guided data synthesis framework tailored for table instruction tuning, named TableDreamer, to mitigate the above issues. Specifically, we first synthesize diverse tables and related instructions as seed data, and then perform an iterative exploration of the input space under the guidance of the newly identified weakness data, which eventually serve as the final training data for fine-tuning the target LLM. Extensive experiments on 10 tabular benchmarks demonstrate the effectiveness of the proposed framework, which boosts the average accuracy of Llama3.1-8B-instruct by 11.62% (49.07% to 60.69%) with 27K GPT-4o synthetic data and outperforms state-of-the-art data synthesis baselines which use more training data. The code and data is available at https://github.com/SpursGoZmy/TableDreamer | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:57:59 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Zheng",
"Mingyu",
""
],
[
"Feng",
"Zhifan",
""
],
[
"Wang",
"Jia",
""
],
[
"Wang",
"Lanrui",
""
],
[
"Lin",
"Zheng",
""
],
[
"Hao",
"Yang",
""
],
[
"Wang",
"Weiping",
""
]
] |
2506.08647 | Oumaima El Khettari | Oumaima El Khettari and Solen Quiniou and Samuel Chaffron | Summarization for Generative Relation Extraction in the Microbiome Domain | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | We explore a generative relation extraction (RE) pipeline tailored to the study of interactions in the intestinal microbiome, a complex and low-resource biomedical domain. Our method leverages summarization with large language models (LLMs) to refine context before extracting relations via instruction-tuned generation. Preliminary results on a dedicated corpus show that summarization improves generative RE performance by reducing noise and guiding the model. However, BERT-based RE approaches still outperform generative models. This ongoing work demonstrates the potential of generative methods to support the study of specialized domains in low-resources setting. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:58:23 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Khettari",
"Oumaima El",
""
],
[
"Quiniou",
"Solen",
""
],
[
"Chaffron",
"Samuel",
""
]
] |
2506.08648 | Lennart Sinjorgo | Lennart Sinjorgo, Renata Sotirov, Juan C. Vera | SDP bounds on the stability number via ADMM and intermediate levels of the Lasserre hierarchy | 17 pages, 9 tables, 1 figure | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the Lasserre hierarchy for computing bounds on the stability number of graphs. The semidefinite programs (SDPs) arising from this hierarchy involve large matrix variables and many linear constraints, which makes them difficult to solve using interior-point methods. We propose solving these SDPs using the alternating direction method of multipliers (ADMM). When the second level of the Lasserre hierarchy for a given graph is intractable for the ADMM, we consider an intermediate-level relaxation of the hierarchy. To warm-start the ADMM, we use an optimal solution from the first level of the Lasserre hierarchy, which is equivalent to the well-known Lov\'asz theta function. Additionally, we use this solution to determine which degree two monomials to add in the Lasserre hierarchy relaxation to obtain an intermediate level between 1 and 2. Computational results demonstrate that our approach yields strong bounds on the stability number, which are computable within reasonable running times. We provide the best-known bounds on the stability number of various graphs from the literature. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 09:58:29 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Sinjorgo",
"Lennart",
""
],
[
"Sotirov",
"Renata",
""
],
[
"Vera",
"Juan C.",
""
]
] |
2506.08649 | Zhiyi Zhu | Zhiyi Zhu, Xiaoyu Wu, Youwei Lu | Enhancing Video Memorability Prediction with Text-Motion Cross-modal Contrastive Loss and Its Application in Video Summarization | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video memorability refers to the ability of videos to be recalled after viewing, playing a crucial role in creating content that remains memorable. Existing models typically focus on extracting multimodal features to predict video memorability scores but often fail to fully utilize motion cues. The representation of motion features is compromised during the fine-tuning phase of the motion feature extractor due to a lack of labeled data. In this paper, we introduce the Text-Motion Cross-modal Contrastive Loss (TMCCL), a multimodal video memorability prediction model designed to enhance the representation of motion features. We tackle the challenge of improving motion feature representation by leveraging text description similarities across videos to establish positive and negative motion sample sets for a given target. This enhancement allows the model to learn similar feature representations for semantically related motion content, resulting in more accurate memorability predictions. Our model achieves state-of-the-art performance on two video memorability prediction datasets. Moreover, the potential applications of video memorability prediction have been underexplored. To address this gap, we present Memorability Weighted Correction for Video Summarization (MWCVS), using video memorability prediction to reduce subjectivity in video summarization labels. Experimental results on two video summarization datasets demonstrate the effectiveness of MWCVS, showcasing the promising applications of video memorability prediction. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:01:08 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Zhu",
"Zhiyi",
""
],
[
"Wu",
"Xiaoyu",
""
],
[
"Lu",
"Youwei",
""
]
] |
2506.08651 | Dina Abdelhadi | Dina Abdelhadi, Colin Sandon, Emmanuel Abbe, Ruediger Urbanke | Reed-Muller Codes for Quantum Pauli and Multiple Access Channels | null | null | null | null | cs.IT math.IT quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reed-Muller (RM) codes have undergone significant analytical advancements over the past decade, particularly for binary memoryless symmetric (BMS) channels. We extend the scope of RM codes development and analysis to multiple-access channels (MACs) and quantum Pauli channels, leveraging a unified approach. Specifically, we first derive the achievable rate region for RM codes on so-called Q-MACs, a class of MACs with additive correlated noise. This is achieved via a generalization of the bending and boosting arguments defined in arXiv:2304.02509. We then put forward a connection between the rate region of these QMACs and quantum RM codes designed for Pauli noise channels. This connection highlights a universality property of quantum RM codes, demonstrating their rate-optimal performance across a range of channel parameters, rather than for a single Pauli channel. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:02:57 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Abdelhadi",
"Dina",
""
],
[
"Sandon",
"Colin",
""
],
[
"Abbe",
"Emmanuel",
""
],
[
"Urbanke",
"Ruediger",
""
]
] |
2506.08652 | Mahesh Godavarti | Mahesh Godavarti | JoFormer (Journey-based Transformer): Theory and Empirical Analysis on the Tiny Shakespeare Dataset | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Transformers have demonstrated remarkable success in sequence modeling, yet effectively incorporating positional information remains a challenging and active area of research. In this paper, we introduce JoFormer, a journey-based Transformer architecture grounded in a recently proposed non-commutative algebra for composing transformations across positions. JoFormer represents relative positions through learnable directional transforms that are sequentially composed along the input, thereby extending and generalizing existing approaches based on relative position representations. We derive the JoFormer attention mechanism from first principles and show that it subsumes standard methods such as rotary transformations as special cases. To evaluate its effectiveness, we compare JoFormer to the RoFormer baseline on the Tiny Shakespeare character-level language modeling task. Our results demonstrate that
JoFormer consistently achieves lower perplexity and faster convergence, highlighting the advantages of its more expressive, journey-based treatment of position. Notably, the per-token JoFormer is still a primitive, conceptual variant with layer-independent angles, yet it already demonstrates strong performance-underscoring its promise as a proof of concept for more expressive architectures. We conclude by discussing how JoFormer offers a principled approach to integrating positional structure into Transformer architectures. The code used in this work is available at https://github.com/mahesh-godavarti/joformer. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:05:29 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Godavarti",
"Mahesh",
""
]
] |
2506.08653 | Alexander Strack | Alexander Strack, Christopher Taylor, and Dirk Pfl\"uger | Parallel FFTW on RISC-V: A Comparative Study including OpenMP, MPI, and HPX | 12 pages, 9 figures, Fifth International workshop on RISC-V for HPC co-located with ISC 2025 | null | null | null | cs.DC cs.ET | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Rapid advancements in RISC-V hardware development shift the focus from low-level optimizations to higher-level parallelization. Recent RISC-V processors, such as the SOPHON SG2042, have 64 cores. RISC-V processors with core counts comparable to the SG2042, make efficient parallelization as crucial for RISC-V as the more established processors such as x86-64. In this work, we evaluate the parallel scaling of the widely used FFTW library on RISC-V for MPI and OpenMP. We compare it to a 64-core AMD EPYC 7742 CPU side by side for different types of FFTW planning. Additionally, we investigate the effect of memory optimization on RISC-V in HPX-FFT, a parallel FFT library based on the asynchronous many-task runtime HPX using an FFTW backend. We generally observe a performance delta between the x86-64 and RISC-V chips of factor eight for double-precision 2D FFT. Effective memory optimizations in HPX-FFT on x86-64 do not translate to the RISC-V chip. FFTW with MPI shows good scaling up to 64 cores on x86-64 and RISC-V regardless of planning. In contrast, FFTW with OpenMP requires measured planning on both architectures to achieve good scaling up to 64 cores. The results of our study mark an early step on the journey to large-scale parallel applications running on RISC-V. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:06:00 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Strack",
"Alexander",
""
],
[
"Taylor",
"Christopher",
""
],
[
"Pflüger",
"Dirk",
""
]
] |
2506.08654 | Ciro Benito Raggio | Ciro Benito Raggio, Paolo Zaffino, Maria Francesca Spadea | A Privacy-Preserving Federated Learning Framework for Generalizable CBCT to Synthetic CT Translation in Head and Neck | null | null | null | null | physics.med-ph cs.LG | http://creativecommons.org/licenses/by/4.0/ | Shortened Abstract
Cone-beam computed tomography (CBCT) has become a widely adopted modality for image-guided radiotherapy (IGRT). However, CBCT suffers from increased noise, limited soft-tissue contrast, and artifacts, resulting in unreliable Hounsfield unit values and hindering direct dose calculation. Synthetic CT (sCT) generation from CBCT addresses these issues, especially using deep learning (DL) methods. Existing approaches are limited by institutional heterogeneity, scanner-dependent variations, and data privacy regulations that prevent multi-center data sharing.
To overcome these challenges, we propose a cross-silo horizontal federated learning (FL) approach for CBCT-to-sCT synthesis in the head and neck region, extending our FedSynthCT framework. A conditional generative adversarial network was collaboratively trained on data from three European medical centers in the public SynthRAD2025 challenge dataset.
The federated model demonstrated effective generalization across centers, with mean absolute error (MAE) ranging from $64.38\pm13.63$ to $85.90\pm7.10$ HU, structural similarity index (SSIM) from $0.882\pm0.022$ to $0.922\pm0.039$, and peak signal-to-noise ratio (PSNR) from $32.86\pm0.94$ to $34.91\pm1.04$ dB. Notably, on an external validation dataset of 60 patients, comparable performance was achieved (MAE: $75.22\pm11.81$ HU, SSIM: $0.904\pm0.034$, PSNR: $33.52\pm2.06$ dB) without additional training, confirming robust generalization despite protocol, scanner differences and registration errors.
These findings demonstrate the technical feasibility of FL for CBCT-to-sCT synthesis while preserving data privacy and offer a collaborative solution for developing generalizable models across institutions without centralized data sharing or site-specific fine-tuning. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:10:56 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Raggio",
"Ciro Benito",
""
],
[
"Zaffino",
"Paolo",
""
],
[
"Spadea",
"Maria Francesca",
""
]
] |
2506.08655 | Kamil Jerabek | Kamil Jerabek, Jan Luxemburk, Richard Plny, Josef Koumar, Jaroslav Pesek, Karel Hynek | When Simple Model Just Works: Is Network Traffic Classification in Crisis? | null | null | null | null | cs.LG cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Machine learning has been applied to network traffic classification (TC) for over two decades. While early efforts used shallow models, the latter 2010s saw a shift toward complex neural networks, often reporting near-perfect accuracy. However, it was recently revealed that a simple k-NN baseline using packet sequences metadata (sizes, times, and directions) can be on par or even outperform more complex methods. In this paper, we investigate this phenomenon further and evaluate this baseline across 12 datasets and 15 TC tasks, and investigate why it performs so well. Our analysis shows that most datasets contain over 50% redundant samples (identical packet sequences), which frequently appear in both training and test sets due to common splitting practices. This redundancy can lead to overestimated model performance and reduce the theoretical maximum accuracy when identical flows have conflicting labels. Given its distinct characteristics, we further argue that standard machine learning practices adapted from domains like NLP or computer vision may be ill-suited for TC. Finally, we propose new directions for task formulation and evaluation to address these challenges and help realign the field. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:11:05 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Jerabek",
"Kamil",
""
],
[
"Luxemburk",
"Jan",
""
],
[
"Plny",
"Richard",
""
],
[
"Koumar",
"Josef",
""
],
[
"Pesek",
"Jaroslav",
""
],
[
"Hynek",
"Karel",
""
]
] |
2506.08656 | Peter Persoon | Peter Persoon | Can knowledge reclassification accelerate technological innovation? | null | null | null | null | econ.GN q-fin.EC | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Technological knowledge evolves not only through the generation of new ideas, but also through the reinterpretation of existing ones. Reinterpretations lead to changes in the classification of knowledge, that is, reclassification. This study investigates how reclassified inventions can serve as renewed sources of innovation, thereby accelerating technological progress. Drawing on patent data as a proxy for technological knowledge, I discuss two empirical patterns: (i) more recent patents are more likely to get reclassified and (ii) larger technological classes acquire proportionally more reclassified patents. Using these patterns, I develop a model that explains how reclassified inventions contribute to faster innovation. The predictions of the model are supported across all major technology domains, suggesting a strong link between reclassification and the pace of technological advancement. More generally, the model connects various, seemingly unrelated knowledge quantities, providing a basis for knowledge intrinsic explanations of growth patterns. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:14:06 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Persoon",
"Peter",
""
]
] |
2506.08658 | Anthea F. Fantina | P. J. Davis, H. Dinh Thi, A. F. Fantina, F. Gulminelli, M. Oertel, L. Suleiman | Crust (Unified) Tool for Equation-of-state Reconstruction (CUTER) v2 | 21 pages, 6 figures. Contribution to the Topical Collection "Nuclear Physics in Astrophysics" in EPJA. arXiv admin note: text overlap with arXiv:2406.14906 | Eur. Phys. J. A 61, 120 (2025) | 10.1140/epja/s10050-025-01562-0 | null | astro-ph.HE nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The equation of state (EoS) is a needed input to determine the neutron-star global properties and to relate them. It is thus important to provide consistent and unified EoSs to avoid possible biases in the analyses coming from the use of inconsistent EoSs. We propose a numerical tool, CUTER, allowing the user to consistently match a nuclear-physics informed crust to an arbitrary higher density EoS. We present here the second version of this tool, CUTER v2. Two functionalities are available with the CUTER v2 tool, allowing the user to reconstruct either the whole (outer and inner) crust, or the outer crust only. We show that the code, that has been tested and validated for use by the astrophysical community, is able to efficiently perform both tasks, allowing the computation of neutron-star global properties in a consistent way. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:19:43 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Davis",
"P. J.",
""
],
[
"Thi",
"H. Dinh",
""
],
[
"Fantina",
"A. F.",
""
],
[
"Gulminelli",
"F.",
""
],
[
"Oertel",
"M.",
""
],
[
"Suleiman",
"L.",
""
]
] |
2506.08659 | Ayaka Shimizu | Yuko Ozawa, Ayaka Shimizu, Yoshiro Yaguchi | The CN matrix of a pure braid projection | 21 pages, 21 figures | null | null | null | math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The CN matrix of an $n$-braid projection $B$ is an $n \times n$ matrix such that each $(i,j)$ entry indicates the number of crossings between $i^{th}$ and $j^{th}$ strands of $B$. In this paper, patterns of an $n \times n$ matrix to be a CN matrix are discussed and the CN matrix of a pure 6-braid projection is characterized. As an application, the OU matrix of a pure 6-braid diagram and the crossing matrix of a positive pure 6-braid are also characterized. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:20:09 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Ozawa",
"Yuko",
""
],
[
"Shimizu",
"Ayaka",
""
],
[
"Yaguchi",
"Yoshiro",
""
]
] |
2506.08660 | Jinkwan Jang | Jinkwan Jang, Hyungjin Park, Jinmyeong Choi, Taesup Kim | Towards Robust Real-World Multivariate Time Series Forecasting: A Unified Framework for Dependency, Asynchrony, and Missingness | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Real-world time series data are inherently multivariate, often exhibiting complex inter-channel dependencies. Each channel is typically sampled at its own period and is prone to missing values due to various practical and operational constraints. These characteristics pose fundamental challenges related to channel dependency, sampling asynchrony, and missingness, all of which must be addressed to enable robust and reliable forecasting in practical settings. However, most existing architectures are built on oversimplified assumptions, such as identical sampling periods across channels and fully observed inputs at test time, which often do not hold in real-world scenarios. To bridge this gap, we propose ChannelTokenFormer, a Transformer-based forecasting model with a flexible architecture designed to explicitly capture cross-channel interactions, accommodate channel-wise asynchronous sampling, and effectively handle missing values. Extensive experiments on three benchmark datasets modified to reflect practical settings, along with one real-world industrial dataset, demonstrate the superior robustness and accuracy of ChannelTokenFormer under challenging real-world conditions. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:21:28 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Jang",
"Jinkwan",
""
],
[
"Park",
"Hyungjin",
""
],
[
"Choi",
"Jinmyeong",
""
],
[
"Kim",
"Taesup",
""
]
] |
2506.08661 | Anya Chaturvedi | Rida Bazzi, Anya Chaturvedi, Andr\'ea W. Richa, Peter Vargas | Synchronization in Anonymous Networks Under Continuous Dynamics | 18 pages, 1 figure, 2 tables | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the $\kappa$-Synchronizer that works in non-synchronous dynamic networks under minimal assumptions. Our model allows continuous topological changes without any guarantee of eventual global or partial stabilization and assumes that nodes are anonymous. This deterministic synchronizer is the first to enable nodes to simulate a dynamic network synchronous algorithm for executions in a semi-synchronous dynamic environment under a weakly-fair node activation scheduler, despite the absence of a global clock, node ids, persistent connectivity or any assumptions about the edge dynamics (in both the synchronous and semi-synchronous environments). In summary, we make the following contributions: (1) we extend the definition of synchronizers to networks with continuous arbitrary edge dynamics; (2) we present the first synchronizer from the semi-synchronous to the synchronous model in a network with continuous arbitrary edge dynamics; and (3) we present non-trivial applications of the proposed synchronizer to existing algorithms. We assume an extension of the Pull communication model by adding a single 1-bit multi-writer atomic register at each edge-port of a node, since we show that the standard Pull model is not sufficient to allow for non-trivial synchronization in our scenario. The $\kappa$-Synchronizer operates with memory overhead at the nodes that is linear on the maximum node degree and logarithmic on the runtime of the underlying synchronous algorithm being simulated. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:22:20 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Bazzi",
"Rida",
""
],
[
"Chaturvedi",
"Anya",
""
],
[
"Richa",
"Andréa W.",
""
],
[
"Vargas",
"Peter",
""
]
] |
2506.08662 | Florian Borzechowski | Florian Borzechowski, Michael Sch\"afer, Heiko Schwarz, Jonathan Pfaff, Detlev Marpe, Thomas Wiegand | Optimizing Learned Image Compression on Scalar and Entropy-Constraint Quantization | Accepted at ICIP2024, the IEEE International Conference on Image Processing | 2024 IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates, 2024, pp. 3688-3694 | 10.1109/ICIP51287.2024.10648254 | null | cs.LG cs.AI cs.IT math.IT | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The continuous improvements on image compression with variational autoencoders have lead to learned codecs competitive with conventional approaches in terms of rate-distortion efficiency. Nonetheless, taking the quantization into account during the training process remains a problem, since it produces zero derivatives almost everywhere and needs to be replaced with a differentiable approximation which allows end-to-end optimization. Though there are different methods for approximating the quantization, none of them model the quantization noise correctly and thus, result in suboptimal networks. Hence, we propose an additional finetuning training step: After conventional end-to-end training, parts of the network are retrained on quantized latents obtained at the inference stage. For entropy-constraint quantizers like Trellis-Coded Quantization, the impact of the quantizer is particularly difficult to approximate by rounding or adding noise as the quantized latents are interdependently chosen through a trellis search based on both the entropy model and a distortion measure. We show that retraining on correctly quantized data consistently yields additional coding gain for both uniform scalar and especially for entropy-constraint quantization, without increasing inference complexity. For the Kodak test set, we obtain average savings between 1% and 2%, and for the TecNick test set up to 2.2% in terms of Bj{\o}ntegaard-Delta bitrate. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:22:22 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Borzechowski",
"Florian",
""
],
[
"Schäfer",
"Michael",
""
],
[
"Schwarz",
"Heiko",
""
],
[
"Pfaff",
"Jonathan",
""
],
[
"Marpe",
"Detlev",
""
],
[
"Wiegand",
"Thomas",
""
]
] |
2506.08663 | Pablo Cova Fari\~na | Pablo Cova Fari\~na, Daniel Jirovec, Xin Zhang, Elizaveta Morozova, Stefan D. Oosterhout, Stefano Reale, Tzu-Kan Hsiao, Giordano Scappucci, Menno Veldhorst, and Lieven M. K. Vandersypen | Site-resolved magnon and triplon dynamics on a programmable quantum dot spin ladder | null | null | null | null | cond-mat.mes-hall quant-ph | http://creativecommons.org/licenses/by/4.0/ | Quasi-particle dynamics in interacting systems in the presence of disorder challenges the notion of internal thermalization, but proves difficult to investigate theoretically for large particle numbers. Engineered quantum systems may offer a viable alternative, as witnessed in experimental demonstrations in a variety of physical platforms, each with its own capabilities and limitations. Semiconductor gate-defined quantum dot arrays are of particular interest since they offer both a direct mapping of their Hamiltonian to Fermi-Hubbard and Heisenberg models and the in-situ tunability of (magnetic) interactions and onsite potentials. In this work, we use an array of germanium quantum dots to simulate the dynamics of both single-spin excitations (magnons) and two-spin excitations (triplons). We develop a methodology that combines digital spin qubit operations for state preparation and readout with analog evolution under the full system Hamiltonian. Using these techniques, we can reconstruct quantum walk plots for both magnons and triplons, and for various configurations of Heisenberg exchange couplings. We furthermore explore the effect of single-site disorder and its impact on the propagation of spin excitations. The obtained results can provide a basis for simulating disorder-based solid-state phenomena such as many-body localization. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:22:43 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Fariña",
"Pablo Cova",
""
],
[
"Jirovec",
"Daniel",
""
],
[
"Zhang",
"Xin",
""
],
[
"Morozova",
"Elizaveta",
""
],
[
"Oosterhout",
"Stefan D.",
""
],
[
"Reale",
"Stefano",
""
],
[
"Hsiao",
"Tzu-Kan",
""
],
[
"Scappucci",
"Giordano",
""
],
[
"Veldhorst",
"Menno",
""
],
[
"Vandersypen",
"Lieven M. K.",
""
]
] |
2506.08664 | Gilles Montambaux | Gilles Montambaux | Ramanujan, Landau and Casimir, divergent series: a physicist point of view | 14 pages, 6 figures, submitted to C. R. Acad. Sci. Paris | null | null | null | cond-mat.other cond-mat.mes-hall | http://creativecommons.org/publicdomain/zero/1.0/ | It is a popular paradoxical exercise to show that the infinite sum of positive integer numbers is equal to -1/12, sometimes called the Ramanujan sum. Here we propose a qualitative approach, much like that of a physicist, to show how the value -1/12 can make sense and, in fact, appears in certain physical quantities where this type of summation is involved. At the light of two physical examples, taken respectively from condensed matter -- the Landau diamagnetism -- and quantum electrodynamics -- the Casimir effect -- that illustrate this strange sum, we present a systematic way to extract this Ramanujan term from the infinity. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:26:04 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Montambaux",
"Gilles",
""
]
] |
2506.08665 | Magnus Ivarsen | Magnus F Ivarsen, Lasse B N Clausen, Yaqi Jin, Jaeheung Park | Meter-scale Observations of Equatorial Plasma Turbulence | 7 pages, 5 figures | null | null | null | physics.space-ph physics.plasm-ph | http://creativecommons.org/licenses/by/4.0/ | The multi-Needle Langmuir Probe collects an electron current through four fixed-bias cylindrical copper needles. This allows for an extremely high sampling frequency, with plasma properties being inferred through polynomial fitting in the current-voltage plane. We present initial results from such a multi-needle probe mounted on the International Space Station, orbiting Earth at an altitude of around 400 km. That altitude, and its orbital inclination (~50 degrees), place the ISS as a suitable platform for observing equatorial plasma bubbles. In case studies of such turbulent structuring of the F-region plasma, we observe density timeseries that conserve considerable detail at virtually every level of magnification down to its Nyquist scale of 2-5 meters. We present power spectral density estimates of the turbulent structuring found inside equatorial plasma bubbles, and we discuss apparent break-points at scale-sizes between 1 m and 300 m, which we interpret in the light of turbulent dissipation as kilometer-scale swirls produced by the gradient-drift instability dissipate in the plasma. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:26:40 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Ivarsen",
"Magnus F",
""
],
[
"Clausen",
"Lasse B N",
""
],
[
"Jin",
"Yaqi",
""
],
[
"Park",
"Jaeheung",
""
]
] |
2506.08666 | Wenzhuo Liu | Wenzhuo Liu, Fei Zhu, Haiyang Guo, Longhui Wei, Cheng-Lin Liu | LLaVA-c: Continual Improved Visual Instruction Tuning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal models like LLaVA-1.5 achieve state-of-the-art visual understanding through visual instruction tuning on multitask datasets, enabling strong instruction-following and multimodal performance. However, multitask learning faces challenges such as task balancing, requiring careful adjustment of data proportions, and expansion costs, where new tasks risk catastrophic forgetting and need costly retraining. Continual learning provides a promising alternative to acquiring new knowledge incrementally while preserving existing capabilities. However, current methods prioritize task-specific performance, neglecting base model degradation from overfitting to specific instructions, which undermines general capabilities. In this work, we propose a simple but effective method with two modifications on LLaVA-1.5: spectral-aware consolidation for improved task balance and unsupervised inquiry regularization to prevent base model degradation. We evaluate both general and task-specific performance across continual pretraining and fine-tuning. Experiments demonstrate that LLaVA-c consistently enhances standard benchmark performance and preserves general capabilities. For the first time, we show that task-by-task continual learning can achieve results that match or surpass multitask joint learning. The code will be publicly released. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:27:52 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Liu",
"Wenzhuo",
""
],
[
"Zhu",
"Fei",
""
],
[
"Guo",
"Haiyang",
""
],
[
"Wei",
"Longhui",
""
],
[
"Liu",
"Cheng-Lin",
""
]
] |
2506.08667 | Prashanta Garain | Gurdev Chand Anthal and Prashanta Garain | Pohozaev-type identities for classes of quasilinear elliptic local and nonlocal equations and systems, with applications | 19 pages, comments are welcome | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this article, we establish Pohozaev-type identities for a class of quasilinear elliptic equations and systems involving both local and nonlocal $p$-Laplace operators. Specifically, we obtain these identities in $\mathbb{R}^n$ for the purely anisotropic $p$-Laplace equations, the purely fractional $p$-Laplace equations, as well as for equations that incorporate both anisotropic and fractional $p$-Laplace features. We also extend these results to the corresponding systems. To the best of our knowledge, the identities we derive in the mixed case are new even when $p=2$. Finally, we illustrate some of the applications of our main results. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:28:53 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Anthal",
"Gurdev Chand",
""
],
[
"Garain",
"Prashanta",
""
]
] |
2506.08668 | Johannes M\"oller | Aliaksandr Leonau, Felix Brausse, James Wrigley, Mads B. Jakobsen, Amir Tosson, Michelle Dargasz, Nimmi Das Anthuparambil, Felix Lehmk\"uhler, Anita Girelli, Maddalena Bin, Fivos Perakis, Sebastian Retzbach, Fajun Zhang, Frank Schreiber, Matheus Teodoro, Cammille Carinan, Robert Rosca, Fabio Dall Antonia, Wonhyuk Jo, Ulrike Boesenberg, Angel Rodriguez-Fernandez, Roman Shayduk, J\"org Hallmann, Alexey Zozulya, Jan-Etienne Pudell, Carsten Deiter, Luca Gelisio, Johannes M\"oller, Anders Madsen, Christian Gutt | A pipeline for Megahertz X-ray Photon Correlation Spectroscopy on soft matter samples at the MID instrument of European XFEL | null | null | null | null | physics.ins-det cond-mat.soft | http://creativecommons.org/licenses/by/4.0/ | In this paper we present the experimental protocol and data processing framework for Megahertz X-ray Photon Correlation Spectroscopy (MHz-XPCS) experiments on soft matter samples, implemented at the Materials Imaging and Dynamics (MID) instrument of the European X-ray Free-Electron Laser (EuXFEL). Due to the introduction of a standard configuration and the implementation of a highly automated data processing pipeline, MHz-XPCS measurements can now be conducted and analyzed with minimal user intervention. A key challenge lies in managing the extremely large data volumes generated by the Adaptive Gain Integrating Pixel Detector (AGIPD) - often reaching several petabytes within a single experiment. We describe the technical implementation, discuss the hardware requirements related to effective parallel data processing, and propose strategies to enhance data quality, in particular related to data reduction strategies and an improvement of the signal-to-noise ratio. Finally, we address strategies for making the processed data FAIR (Findable, Accessible, Interoperable, Reusable), in alignment with the goals of the DAPHNE4NFDI project. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:30:40 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Leonau",
"Aliaksandr",
""
],
[
"Brausse",
"Felix",
""
],
[
"Wrigley",
"James",
""
],
[
"Jakobsen",
"Mads B.",
""
],
[
"Tosson",
"Amir",
""
],
[
"Dargasz",
"Michelle",
""
],
[
"Anthuparambil",
"Nimmi Das",
""
],
[
"Lehmkühler",
"Felix",
""
],
[
"Girelli",
"Anita",
""
],
[
"Bin",
"Maddalena",
""
],
[
"Perakis",
"Fivos",
""
],
[
"Retzbach",
"Sebastian",
""
],
[
"Zhang",
"Fajun",
""
],
[
"Schreiber",
"Frank",
""
],
[
"Teodoro",
"Matheus",
""
],
[
"Carinan",
"Cammille",
""
],
[
"Rosca",
"Robert",
""
],
[
"Antonia",
"Fabio Dall",
""
],
[
"Jo",
"Wonhyuk",
""
],
[
"Boesenberg",
"Ulrike",
""
],
[
"Rodriguez-Fernandez",
"Angel",
""
],
[
"Shayduk",
"Roman",
""
],
[
"Hallmann",
"Jörg",
""
],
[
"Zozulya",
"Alexey",
""
],
[
"Pudell",
"Jan-Etienne",
""
],
[
"Deiter",
"Carsten",
""
],
[
"Gelisio",
"Luca",
""
],
[
"Möller",
"Johannes",
""
],
[
"Madsen",
"Anders",
""
],
[
"Gutt",
"Christian",
""
]
] |
2506.08669 | Dongge Han | Dongge Han, Menglin Xia, Daniel Madrigal Diaz, Samuel Kessler, Ankur Mallick, Xuchao Zhang, Mirian Del Carmen Hipolito Garcia, Jin Xu, Victor R\"uhle, Saravan Rajmohan | Enhancing Reasoning Capabilities of Small Language Models with Blueprints and Prompt Template Search | TTODLer-FM Workshop@ICML'25 (Tiny Titans: The next wave of On-Device Learning for Foundational Models) | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Small language models (SLMs) offer promising and efficient alternatives to large language models (LLMs). However, SLMs' limited capacity restricts their reasoning capabilities and makes them sensitive to prompt variations. To address these challenges, we propose a novel framework that enhances SLM reasoning capabilities through LLM generated blueprints. The blueprints provide structured, high-level reasoning guides that help SLMs systematically tackle related problems. Furthermore, our framework integrates a prompt template search mechanism to mitigate the SLMs' sensitivity to prompt variations. Our framework demonstrates improved SLM performance across various tasks, including math (GSM8K), coding (MBPP), and logic reasoning (BBH). Our approach improves the reasoning capabilities of SLMs without increasing model size or requiring additional training, offering a lightweight and deployment-friendly solution for on-device or resource-constrained environments. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:30:43 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Han",
"Dongge",
""
],
[
"Xia",
"Menglin",
""
],
[
"Diaz",
"Daniel Madrigal",
""
],
[
"Kessler",
"Samuel",
""
],
[
"Mallick",
"Ankur",
""
],
[
"Zhang",
"Xuchao",
""
],
[
"Garcia",
"Mirian Del Carmen Hipolito",
""
],
[
"Xu",
"Jin",
""
],
[
"Rühle",
"Victor",
""
],
[
"Rajmohan",
"Saravan",
""
]
] |
2506.08670 | Renjie Xu | Renjie Xu, Chong Wu, Maolin Che, Zhuoheng Ran, Yimin Wei and Hong Yan | sparseGeoHOPCA: A Geometric Solution to Sparse Higher-Order PCA Without Covariance Estimation | null | null | null | null | math.NA cs.LG cs.NA math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose sparseGeoHOPCA, a novel framework for sparse higher-order principal component analysis (SHOPCA) that introduces a geometric perspective to high-dimensional tensor decomposition. By unfolding the input tensor along each mode and reformulating the resulting subproblems as structured binary linear optimization problems, our method transforms the original nonconvex sparse objective into a tractable geometric form. This eliminates the need for explicit covariance estimation and iterative deflation, enabling significant gains in both computational efficiency and interpretability, particularly in high-dimensional and unbalanced data scenarios. We theoretically establish the equivalence between the geometric subproblems and the original SHOPCA formulation, and derive worst-case approximation error bounds based on classical PCA residuals, providing data-dependent performance guarantees. The proposed algorithm achieves a total computational complexity of $O\left(\sum_{n=1}^{N} (k_n^3 + J_n k_n^2)\right)$, which scales linearly with tensor size. Extensive experiments demonstrate that sparseGeoHOPCA accurately recovers sparse supports in synthetic settings, preserves classification performance under 10$\times$ compression, and achieves high-quality image reconstruction on ImageNet, highlighting its robustness and versatility. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:30:48 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Xu",
"Renjie",
""
],
[
"Wu",
"Chong",
""
],
[
"Che",
"Maolin",
""
],
[
"Ran",
"Zhuoheng",
""
],
[
"Wei",
"Yimin",
""
],
[
"Yan",
"Hong",
""
]
] |
2506.08671 | Junfeng Liu | Junfeng Liu, Jiarui Ye, Mengshi Chen, Meng Li, Siqiang Luo | Evaluating Learned Indexes in LSM-tree Systems: Benchmarks,Insights and Design Choices | 14 pages,12 figures | null | null | null | cs.DB cs.DS | http://creativecommons.org/licenses/by/4.0/ | LSM-tree-based data stores are widely used in industry due to their exceptional performance. However, as data volumes grow, efficiently querying large-scale databases becomes increasingly challenging. To address this, recent studies attempted to integrate learned indexes into LSM-trees to enhance lookup performance, which has demonstrated promising improvements. Despite this, only a limited range of learned index types has been considered, and the strengths and weaknesses of different learned indexes remain unclear, making them difficult for practical use. To fill this gap, we provide a comprehensive and systematic benchmark to pursue an in-depth understanding of learned indexes in LSM-tree systems. In this work, we summarize the workflow of 8 existing learned indexes and analyze the associated theoretical cost. We also identify several key factors that significantly influence the performance of learned indexes and conclude them with a novel configuration space, including various index types, boundary positions, and granularity. Moreover, we implement different learned index designs on a unified platform to evaluate across various configurations. Surprisingly, our experiments reveal several unexpected insights, such as the marginal lookup enhancement when allocating a large memory budget to learned indexes and modest retraining overhead of learned indexes. Besides, we also offer practical guidelines to help developers intelligently select and tune learned indexes for custom use cases. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:31:09 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Liu",
"Junfeng",
""
],
[
"Ye",
"Jiarui",
""
],
[
"Chen",
"Mengshi",
""
],
[
"Li",
"Meng",
""
],
[
"Luo",
"Siqiang",
""
]
] |
2506.08675 | Si-Yuan Zhu | Si-Yuan Zhu, Hui-Ying Deng, Fu-Wen Zhang, Qian-Zi Mo, Pak-Hin Thomas Tam | Identifying Merger-Driven Long Gamma-Ray Bursts based on Machine Learning | 13 pages, 7 figures, 5 tables, Accepted for publication in MNRAS | null | null | null | astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Gamma-ray bursts (GRBs) are classified as Type I GRBs originated from compact binary mergers and Type II GRBs originated from massive collapsars. While Type I GRBs are typically shorter than 2 seconds, recent observations suggest that some extend to tens of seconds, forming a potential subclass, Type IL GRBs. However, apart from their association with kilonovae, so far no rapid identification is possible. Given the uncertainties and limitations of optical and infrared afterglow observations, an identification method based solely on prompt emission can make such identification possible for many more GRBs. Interestingly, two established Type IL GRBs: GRB 211211A and GRB 230307A, exhibit a three-episode structure: precursor emission (PE), main emission (ME), and extended emission. Therefore, we comprehensively search for GRBs in the Fermi/GBM catalog and identify 29 three-episode GRBs. Based on 12 parameters, we utilize machine learning to distinguish Type IL GRBs from Type II GRBs. Apart from GRB 211211A and GRB 230307A, we are able to identify six more previously unknown Type IL GRBs: GRB 090831, GRB 170228A, GRB 180605A, GRB 200311A, GRB 200914A, and GRB 211019A. We find that Type IL GRBs are characterized by short duration and minimum variability timescale of PE, a short waiting time between PE and ME, and that ME follows the $E_{\rm p,z}$--$E_{\rm iso}$ correlation of Type I GRBs. For the first time, we identify a high-significant PE in the confirmed Type IL GRB 060614. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:34:19 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Zhu",
"Si-Yuan",
""
],
[
"Deng",
"Hui-Ying",
""
],
[
"Zhang",
"Fu-Wen",
""
],
[
"Mo",
"Qian-Zi",
""
],
[
"Tam",
"Pak-Hin Thomas",
""
]
] |
2506.08676 | Gregorio I. Sainz-Palmero | Alicia Beneyto-Rodriguez, Gregorio I. Sainz-Palmero, Marta Galende-Hern\'andez and Mar\'ia J. Fuente | Linguistic Ordered Weighted Averaging based deep learning pooling for fault diagnosis in a wastewater treatment plant | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Nowadays, water reuse is a serious challenge to help address water shortages. Here, the wastewater treatment plants (WWTP) play a key role, and its proper operation is mandatory. So, fault diagnosis is a key activity for these plants. Their high complexity and large-scale require of smart methodologies for that fault diagnosis and safety operation. All these large-scale and complex industrial processes are monitored, allowing the data collection about the plant operation, so data driven approaches for fault diagnosis can be applied. A popular approach to fault diagnosis is deep learning-based methodologies. Here, a fault diagnosis methodology is proposed for a WWTP using a new linguistic Ordered Weighted Averaging (OWA) pooling based Deep Convolutional Neural Network (DCNN) and a sliding and overlapping time window. This window slides over input data based on the monitoring sampling time, then the diagnosis is carried out by the linguistic OWA pooling based DCNN. This alternative linguistic pooling uses well-known linguistic OWA quantifiers, which permit terms such as \textsl{Most, AtLeast, etc.}, supplying new intuitive options for the pooling tasks. This sliding time window and the OWA pooling based network permit a better and earlier fault diagnosis, at each sampling time, using a few monitoring samples and a fewer learning iterations than DCNN standard pooling. Several linguistic OWA operators have been checked with a benchmark for WWTPs. A set of 5 fault types has been used, taking into account 140 variables sampled at 15 minutes time intervals. The performance has been over $91\%$ for $Accuracy$, $Recall$ or $F1-Score$, and better than other competitive methodologies. Moreover, these linguistic OWA operators for DCNN pooling have shown a better performance than the standard \textsl{Max} and \textsl{Average} options. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:35:56 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Beneyto-Rodriguez",
"Alicia",
""
],
[
"Sainz-Palmero",
"Gregorio I.",
""
],
[
"Galende-Hernández",
"Marta",
""
],
[
"Fuente",
"María J.",
""
]
] |
2506.08677 | Milica \v{S}kipina | Milica \v{S}kipina, Nikola Jovi\v{s}i\'c, Nicola Dall'Asen, Vanja \v{S}venda, Anil Osman Tur, Slobodan Ili\'c, Elisa Ricci, Dubravko \'Culibrk | MAMBO: High-Resolution Generative Approach for Mammography Images | 21 pages, 14 figures, 7 tables | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Mammography is the gold standard for the detection and diagnosis of breast cancer. This procedure can be significantly enhanced with Artificial Intelligence (AI)-based software, which assists radiologists in identifying abnormalities. However, training AI systems requires large and diverse datasets, which are often difficult to obtain due to privacy and ethical constraints. To address this issue, the paper introduces MAMmography ensemBle mOdel (MAMBO), a novel patch-based diffusion approach designed to generate full-resolution mammograms. Diffusion models have shown breakthrough results in realistic image generation, yet few studies have focused on mammograms, and none have successfully generated high-resolution outputs required to capture fine-grained features of small lesions. To achieve this, MAMBO integrates separate diffusion models to capture both local and global (image-level) contexts. The contextual information is then fed into the final patch-based model, significantly aiding the noise removal process. This thoughtful design enables MAMBO to generate highly realistic mammograms of up to 3840x3840 pixels. Importantly, this approach can be used to enhance the training of classification models and extended to anomaly detection. Experiments, both numerical and radiologist validation, assess MAMBO's capabilities in image generation, super-resolution, and anomaly detection, highlighting its potential to enhance mammography analysis for more accurate diagnoses and earlier lesion detection. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:37:21 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Škipina",
"Milica",
""
],
[
"Jovišić",
"Nikola",
""
],
[
"Dall'Asen",
"Nicola",
""
],
[
"Švenda",
"Vanja",
""
],
[
"Tur",
"Anil Osman",
""
],
[
"Ilić",
"Slobodan",
""
],
[
"Ricci",
"Elisa",
""
],
[
"Ćulibrk",
"Dubravko",
""
]
] |
2506.08678 | Juan Yeo | Juan Yeo, Soonwoo Cha, Jiwoo Song, Hyunbin Jin, Taesup Kim | ATAS: Any-to-Any Self-Distillation for Enhanced Open-Vocabulary Dense Prediction | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Vision-language models such as CLIP have recently propelled open-vocabulary dense prediction tasks by enabling recognition of a broad range of visual concepts. However, CLIP still struggles with fine-grained, region-level understanding, hindering its effectiveness on these dense prediction tasks. We identify two pivotal factors required to address this limitation: semantic coherence and fine-grained vision-language alignment. Current adaptation methods often improve fine-grained alignment at the expense of semantic coherence, and often rely on extra modules or supervised fine-tuning. To overcome these issues, we propose Any-to-Any Self-Distillation (ATAS), a novel approach that simultaneously enhances semantic coherence and fine-grained alignment by leveraging own knowledge of a model across all representation levels. Unlike prior methods, ATAS uses only unlabeled images and an internal self-distillation process to refine representations of CLIP vision encoders, preserving local semantic consistency while sharpening local detail recognition. On open-vocabulary object detection and semantic segmentation benchmarks, ATAS achieves substantial performance gains, outperforming baseline CLIP models. These results validate the effectiveness of our approach and underscore the importance of jointly maintaining semantic coherence and fine-grained alignment for advanced open-vocabulary dense prediction. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:40:10 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Yeo",
"Juan",
""
],
[
"Cha",
"Soonwoo",
""
],
[
"Song",
"Jiwoo",
""
],
[
"Jin",
"Hyunbin",
""
],
[
"Kim",
"Taesup",
""
]
] |
2506.08679 | Waldemar Schefer | Waldemar Schefer | On the domains of first order differential operators on the Sierpi\'{n}ski gasket | 36 pages, 3 figures | null | null | null | math.FA math.AP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We study the first order structure of $L^2$-differential one-forms on the Sierpi\'{n}ski gasket. We consider piecewise energy finite functions related to one-forms, normal parts, and show self-similarity properties of one-forms as in the case of energy finite functions on the Sierpi\'{n}ski gasket. We introduce first order differential operators, taking functions into functions, that can be understood as total derivatives with respect to some reference function or form. The main result is a pointwise representation result for certain elements in the domain of said first order differential operator by a ratio of normal derivatives. At last we provide three classes of examples in the domain of said first order differential operator with different types of discontinuities. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:43:31 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Schefer",
"Waldemar",
""
]
] |
2506.08680 | Hugo Daniel Macedo | Hugo Daniel Macedo, Ken Pierce | Proceedings of the 23rd International Overture Workshop | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This volume contains the papers presented at the 23rd International Overture Workshop, held on the 11th of June 2025. This event was the latest in a series of workshops around the Vienna Development Method (VDM), the open-source project Overture, and related tools and formalisms. VDM is one of the longest established formal methods for systems development. A lively community of researchers and practitioners has grown up in academia and industry has grown around the modelling languages (VDM-SL, VDM++, VDM-RT, CML) and tools (VDMTools, Overture, Crescendo, Symphony, the INTO-CPS chain, and ViennaTalk). Together, these provide a platform for work on modelling and analysis technology that includes static and dynamic analysis, test generation, execution support, and model checking. This workshop provided updates on the emerging technology of VDM/Overture, including collaboration infrastructure, collaborative modelling and co-simulation for Cyber-Physical Systems. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:44:47 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Macedo",
"Hugo Daniel",
""
],
[
"Pierce",
"Ken",
""
]
] |
2506.08682 | Hector Puerta Ramisa | Tobias Hansen, Paul Heslop, Hector Puerta Ramisa | Superconformal Weight Shifting Operators | 84 pages | null | null | null | hep-th math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | We develop a framework for constructing superconformal blocks for correlators of general supermultiplets in theories with $SU(m,m|2n)$ symmetry, such as four-dimensional $\mathcal{N}\!=\!2$ and $\mathcal{N} \!=\! 4$ conformal theories. We use analytic superspace, viewed as the super-Grassmannian $\mathrm{Gr}(m|n,2m|2n)$, which includes 4D Minkowski space ($m=2,n=0$). In this formalism, superblocks for non-half-BPS correlators are analogous to non-supersymmetric conformal blocks for correlators of fields with spin. We construct $SU(m,m|2n)$-covariant differential operators which generalise the existing conformal weight-shifting operators, and thus allow us to derive all superconformal blocks from the known half-BPS blocks. Our results provide a framework from which to advance the conformal bootstrap in 4D supersymmetric settings, with potential extensions to lower and higher-dimensional SCFTs. The Grassmannian formalism is also seen to offer a natural and often simpler alternative to the embedding space formalism of non-supersymmetric CFTs. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:46:21 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Hansen",
"Tobias",
""
],
[
"Heslop",
"Paul",
""
],
[
"Ramisa",
"Hector Puerta",
""
]
] |
2506.08683 | Yakov Braver | Y. Braver, D. Burba, S. S. Nair, G. \v{Z}labys, E. Anisimovas, Th. Busch, G. Juzeli\=unas | Light-induced localized vortices in multicomponent Bose-Einstein condensates | 11 pages, 6 figures | null | null | null | cond-mat.quant-gas | http://creativecommons.org/licenses/by/4.0/ | We study continuous interaction of a trapped two-component Bose-Einstein condensate with light fields in a $\Lambda$-type configuration. Using light beams with orbital angular momentum, we theoretically show how to create a stable, pinned vortex configuration, where the rotating component is confined to the region surrounded by the second, non-rotating component. The atoms constituting this vortex can be localized in volumes much smaller than the volume occupied by the second component. The position of the vortex can be robustly changed by moving the laser beams as long as the beam movement speed is below the speed of sound. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:49:28 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Braver",
"Y.",
""
],
[
"Burba",
"D.",
""
],
[
"Nair",
"S. S.",
""
],
[
"Žlabys",
"G.",
""
],
[
"Anisimovas",
"E.",
""
],
[
"Busch",
"Th.",
""
],
[
"Juzeliūnas",
"G.",
""
]
] |
2506.08684 | James Tener | Andr\'e G. Henriques and James E. Tener | Integrating positive energy representations of the Virasoro algebra | 32 pages | null | null | null | math.FA math-ph math.MP math.RT | http://creativecommons.org/licenses/by/4.0/ | We show that every unitary positive energy representation W of the Virasoro algebra exponentiates to a holomorphic *-representation of the semigroup of annuli by bounded operators on the Hilbert space completion of W. We use this to show that every representation of the Virasoro conformal net also carries a representation of the semigroup of annuli of the same kind. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:50:11 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Henriques",
"André G.",
""
],
[
"Tener",
"James E.",
""
]
] |
2506.08685 | Liping Li | Zhenxing Di, Liping Li, and Li Liang | A torsion theoretic interpretation for sheaves of modules and Grothendieck topologies on directed categories | null | null | null | null | math.RT math.CT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We prove that every Grothendieck topology induces a hereditary torsion pair in the category of presheaves of modules on a ringed site, and obtain a homological characterization of sheaves of modules: a presheaf of modules is a sheaf of modules if and only if it is saturated with respect to torsion presheaves, or equivalently, it is right perpendicular to torsion presheaves in the sense of Geigle and Lenzing. We also study Grothendieck topologies on directed categories $\mathscr{C}$, and show that every Grothendieck topology on $\mathscr{C}$ is a subcategory topology if and only if $\mathscr{C}$ is an artinian EI category. Consequently, in this case every sheaf category is equivalent to the presheaf category over a full subcategory of $\mathscr{C}$. Finally, we give a few potential applications of these results in group representation theory. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:50:36 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Di",
"Zhenxing",
""
],
[
"Li",
"Liping",
""
],
[
"Liang",
"Li",
""
]
] |
2506.08686 | Soham Poddar | Soham Poddar, Paramita Koley, Janardan Misra, Sanjay Podder, Navveen Balani, Niloy Ganguly, Saptarshi Ghosh | Brevity is the soul of sustainability: Characterizing LLM response lengths | Accepted to appear at the ACL 2025 findings | null | null | null | cs.CL cs.CY | http://creativecommons.org/licenses/by/4.0/ | A significant portion of the energy consumed by Large Language Models (LLMs) arises from their inference processes; hence developing energy-efficient methods for inference is crucial. While several techniques exist for inference optimization, output compression remains relatively unexplored, with only a few preliminary efforts addressing this aspect. In this work, we first benchmark 12 decoder-only LLMs across 5 datasets, revealing that these models often produce responses that are substantially longer than necessary. We then conduct a comprehensive quality assessment of LLM responses, formally defining six information categories present in LLM responses. We show that LLMs often tend to include redundant or additional information besides the minimal answer. To address this issue of long responses by LLMs, we explore several simple and intuitive prompt-engineering strategies. Empirical evaluation shows that appropriate prompts targeting length reduction and controlling information content can achieve significant energy optimization between 25-60\% by reducing the response length while preserving the quality of LLM responses. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:52:04 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Poddar",
"Soham",
""
],
[
"Koley",
"Paramita",
""
],
[
"Misra",
"Janardan",
""
],
[
"Podder",
"Sanjay",
""
],
[
"Balani",
"Navveen",
""
],
[
"Ganguly",
"Niloy",
""
],
[
"Ghosh",
"Saptarshi",
""
]
] |
2506.08687 | Lingjuan Shi | Chengqi Li, Jinhui Yin, Lingjuan Shi | The number of maximal matchings in polygon rings | 21 pages,5 figures,2 tables | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A matching of graph $G$ is maximal if it cannot be expanded by adding any edge to create a larger matching. In this paper, for a hexagonal ring $H$ with $n$ hexagons, we show that the number of maximal matchings of $H$ equals to the trace of the product of $n$ matrices, each of which is $S$, $L$, or $R$ according to the type of the connection mode of $H$. Finally, we extend this conclusion to arbitrary polygon rings and provide an algorithm to determine the transition matrices of polygon chains (rings). | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:53:14 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Li",
"Chengqi",
""
],
[
"Yin",
"Jinhui",
""
],
[
"Shi",
"Lingjuan",
""
]
] |
2506.08688 | Wenbing Tang | Wenbing Tang, Mingfei Cheng, Renzhi Wang, Yuan Zhou, Chengwei Liu, Yang Liu, Zuohua Ding | Causality-aware Safety Testing for Autonomous Driving Systems | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Simulation-based testing is essential for evaluating the safety of Autonomous Driving Systems (ADSs). Comprehensive evaluation requires testing across diverse scenarios that can trigger various types of violations under different conditions. While existing methods typically focus on individual diversity metrics, such as input scenarios, ADS-generated motion commands, and system violations, they often fail to capture the complex interrelationships among these elements. This oversight leads to gaps in testing coverage, potentially missing critical issues in the ADS under evaluation. However, quantifying these interrelationships presents a significant challenge. In this paper, we propose a novel causality-aware fuzzing technique, Causal-Fuzzer, to enable efficient and comprehensive testing of ADSs by exploring causally diverse scenarios. The core of Causal-Fuzzer is constructing a causal graph to model the interrelationships among the diversities of input scenarios, ADS motion commands, and system violations. Then the causal graph will guide the process of critical scenario generation. Specifically, Causal-Fuzzer proposes (1) a causality-based feedback mechanism that quantifies the combined diversity of test scenarios by assessing whether they activate new causal relationships, and (2) a causality-driven mutation strategy that prioritizes mutations on input scenario elements with higher causal impact on ego action changes and violation occurrence, rather than treating all elements equally. We evaluated Causal-Fuzzer on an industry-grade ADS Apollo, with a high-fidelity. Our empirical results demonstrate that Causal-Fuzzer significantly outperforms existing methods in (1) identifying a greater diversity of violations, (2) providing enhanced testing sufficiency with improved coverage of causal relationships, and (3) achieving greater efficiency in detecting the first critical scenarios. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:53:26 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Tang",
"Wenbing",
""
],
[
"Cheng",
"Mingfei",
""
],
[
"Wang",
"Renzhi",
""
],
[
"Zhou",
"Yuan",
""
],
[
"Liu",
"Chengwei",
""
],
[
"Liu",
"Yang",
""
],
[
"Ding",
"Zuohua",
""
]
] |
2506.08690 | Hugo Porta | Hugo Porta, Emanuele Dalsasso, Jessica L. McCarty, Devis Tuia | CanadaFireSat: Toward high-resolution wildfire forecasting with multiple modalities | 34 pages, 11 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Canada experienced in 2023 one of the most severe wildfire seasons in recent history, causing damage across ecosystems, destroying communities, and emitting large quantities of CO2. This extreme wildfire season is symptomatic of a climate-change-induced increase in the length and severity of the fire season that affects the boreal ecosystem. Therefore, it is critical to empower wildfire management in boreal communities with better mitigation solutions. Wildfire probability maps represent an important tool for understanding the likelihood of wildfire occurrence and the potential severity of future wildfires. The massive increase in the availability of Earth observation data has enabled the development of deep learning-based wildfire forecasting models, aiming at providing precise wildfire probability maps at different spatial and temporal scales. A main limitation of such methods is their reliance on coarse-resolution environmental drivers and satellite products, leading to wildfire occurrence prediction of reduced resolution, typically around $\sim 0.1${\deg}. This paper presents a benchmark dataset: CanadaFireSat, and baseline methods for high-resolution: 100 m wildfire forecasting across Canada, leveraging multi-modal data from high-resolution multi-spectral satellite images (Sentinel-2 L1C), mid-resolution satellite products (MODIS), and environmental factors (ERA5 reanalysis data). Our experiments consider two major deep learning architectures. We observe that using multi-modal temporal inputs outperforms single-modal temporal inputs across all metrics, achieving a peak performance of 60.3% in F1 score for the 2023 wildfire season, a season never seen during model training. This demonstrates the potential of multi-modal deep learning models for wildfire forecasting at high-resolution and continental scale. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:58:43 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Porta",
"Hugo",
""
],
[
"Dalsasso",
"Emanuele",
""
],
[
"McCarty",
"Jessica L.",
""
],
[
"Tuia",
"Devis",
""
]
] |
2506.08691 | Congzhi Zhang | Congzhi Zhang, Jiawei Peng, Zhenglin Wang, Yilong Lai, Haowen Sun, Heng Chang, Fei Ma, Weijiang Yu | VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism | Accepted by ACL 2025 main | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Large Vision-Language Models (LVLMs) have shown exceptional performance in multimodal tasks, but their effectiveness in complex visual reasoning is still constrained, especially when employing Chain-of-Thought prompting techniques. In this paper, we propose VReST, a novel training-free approach that enhances Reasoning in LVLMs through Monte Carlo Tree Search and Self-Reward mechanisms. VReST meticulously traverses the reasoning landscape by establishing a search tree, where each node encapsulates a reasoning step, and each path delineates a comprehensive reasoning sequence. Our innovative multimodal Self-Reward mechanism assesses the quality of reasoning steps by integrating the utility of sub-questions, answer correctness, and the relevance of vision-language clues, all without the need for additional models. VReST surpasses current prompting methods and secures state-of-the-art performance across three multimodal mathematical reasoning benchmarks. Furthermore, it substantiates the efficacy of test-time scaling laws in multimodal tasks, offering a promising direction for future research. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 11:02:36 GMT"
}
] | 2025-06-11T00:00:00 | [
[
"Zhang",
"Congzhi",
""
],
[
"Peng",
"Jiawei",
""
],
[
"Wang",
"Zhenglin",
""
],
[
"Lai",
"Yilong",
""
],
[
"Sun",
"Haowen",
""
],
[
"Chang",
"Heng",
""
],
[
"Ma",
"Fei",
""
],
[
"Yu",
"Weijiang",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.