id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-04-25 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.14194 | Xinlin Zhuang | Xinlin Zhuang, Jiahui Peng, Ren Ma, Yinfan Wang, Tianyi Bai, Xingjian
Wei, Jiantao Qiu, Chi Zhang, Ying Qian, Conghui He | Meta-rater: A Multi-dimensional Data Selection Method for Pre-training
Language Models | Under review | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The composition of pre-training datasets for large language models (LLMs)
remains largely undisclosed, hindering transparency and efforts to optimize
data quality, a critical driver of model performance. Current data selection
methods, such as natural language quality assessments, diversity-based filters,
and classifier-based approaches, are limited by single-dimensional evaluation
or redundancy-focused strategies. To address these gaps, we propose PRRC to
evaluate data quality across Professionalism, Readability, Reasoning, and
Cleanliness. We further introduce Meta-rater, a multi-dimensional data
selection method that integrates these dimensions with existing quality metrics
through learned optimal weightings. Meta-rater employs proxy models to train a
regression model that predicts validation loss, enabling the identification of
optimal combinations of quality scores. Experiments demonstrate that Meta-rater
doubles convergence speed for 1.3B parameter models and improves downstream
task performance by 3.23, with scalable benefits observed in 3.3B models
trained on 100B tokens. Additionally, we release the annotated SlimPajama-627B
dataset, labeled across 25 quality metrics (including PRRC), to advance
research in data-centric LLM development. Our work establishes that holistic,
multi-dimensional quality integration significantly outperforms conventional
single-dimension approaches, offering a scalable paradigm for enhancing
pre-training efficiency and model capability.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:12:33 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhuang",
"Xinlin",
""
],
[
"Peng",
"Jiahui",
""
],
[
"Ma",
"Ren",
""
],
[
"Wang",
"Yinfan",
""
],
[
"Bai",
"Tianyi",
""
],
[
"Wei",
"Xingjian",
""
],
[
"Qiu",
"Jiantao",
""
],
[
"Zhang",
"Chi",
""
],
[
"Qian",
"Ying",
""
],
[
"He",
"Conghui",
""
]
] |
2504.14195 | Michelle D\"oring | Michelle D\"oring, Markus Brill and Jobst Heitzig | The River Method | null | null | null | null | cs.GT | http://creativecommons.org/licenses/by/4.0/ | We introduce River, a novel Condorcet-consistent voting method that is based
on pairwise majority margins and can be seen as a simplified variation of
Tideman's Ranked Pairs method. River is simple to explain, simple to compute
even 'by hand', and gives rise to an easy-to-interpret certificate in the form
of a directed tree. Like Ranked Pairs and Schulze's Beat Path method, River is
a refinement of the Split Cycle method and shares with those many desirable
properties, including independence of clones. Unlike the other three methods,
River satisfies a strong form of resistance to agenda-manipulation that is
known as independence of Pareto-dominated alternatives.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:13:26 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Döring",
"Michelle",
""
],
[
"Brill",
"Markus",
""
],
[
"Heitzig",
"Jobst",
""
]
] |
2504.14196 | Xiaojun Wu | Deyin Kong, Yichen Su, Cheng Song, Xiaojun Wu | Room-temperature high-average-power strong-field terahertz source based
on industrial high-repetition-rate femtosecond laser | 15 pages, 9 figures | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Free-space strong-field terahertz (THz) pulses, generated via optical
rectification of femtosecond lasers in nonlinear crystals, are pivotal in
various applications. However, conventional Ti:sapphire lasers struggle to
produce high-average-power THz due to their limited output power. While
kilowatt ytterbium lasers are increasingly adopted, their application in THz
generation faces challenges: low optical-to-THz conversion efficiency
(attributed to long pulse durations and low energy) and crystal damage under
high pumping power. Here, we report a high-average-power strong-field THz
source using a lithium niobate crystal pumped by a 1030-nm, 570-fs, 1-mJ,
50-kHz ytterbium femtosecond laser with tilted pulse front pumping (TPFP). By
systematically optimizing TPFP implementations and comparing grating- and
echelon-type configurations, we achieve a THz source with 64.5 mW average power
at 42-W, 50-kHz pumping, and a focused peak electric field of 525 kV/cm at
0.83-mJ, 1-kHz operation. Additionally, we observe Zeeman torque signals in
cobalt-iron ferromagnetic nanofilms. This high-repetition-rate,
high-average-power THz system, combined with its potential capabilities in high
signal-to-noise spectroscopy and imaging, promises transformative impacts in
quantum matter manipulation, non-destructive testing, and biomedicine.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:18:56 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Kong",
"Deyin",
""
],
[
"Su",
"Yichen",
""
],
[
"Song",
"Cheng",
""
],
[
"Wu",
"Xiaojun",
""
]
] |
2504.14197 | Jameel-Un Nabi | Jameel-Un Nabi, Ramoona Shehzadi, Muhammad Majid | Neutrino cooling rates due to nickel isotopes for presupernova evolution
of massive stars | 33 pages, 8 tables, 9 figures | New Astronomy, 71, 12-24 (2019) | 10.1016/j.newast.2019.03.001 | null | nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Simulation studies indicate that weak interaction rates on nickel isotopes
play a crucial role in determining the electron-to-baryon ratio within the
stellar interior during the late stages of core evolution. (Anti)neutrinos
produced through weak decay processes escape from stellar regions with
densities below 10^11 g/cm^3, carrying away energy and thereby reducing the
core entropy. In this work, we present a microscopic calculation of neutrino
and antineutrino cooling rates resulting from weak interactions on nickel
isotopes in the mass range 56 <= A <= 71. The calculations are performed using
the deformed proton-neutron Quasiparticle Random Phase Approximation (pn-QRPA)
model. Recent investigations into the Gamow-Teller (GT) strength distributions
of nickel isotopes demonstrate that the deformed pn-QRPA model successfully
reproduces experimental charge-changing transition data.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:18:57 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Nabi",
"Jameel-Un",
""
],
[
"Shehzadi",
"Ramoona",
""
],
[
"Majid",
"Muhammad",
""
]
] |
2504.14198 | Tejbir Lohan | Chi-Kwong Li, Tejbir Lohan, and Sushil Singla | Linear maps preserving product of involutions | 22 pages, comments are welcome! | null | null | null | math.FA math.GR math.OA math.RA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | An element of the algebra $M_n(\mathbb{F})$ of $n \times n$ matrices over a
field $\mathbb{F}$ is called an involution if its square equals the identity
matrix. Gustafson, Halmos, and Radjavi proved that any product of involutions
in $M_n(\mathbb{F})$ can be expressed as a product of at most four involutions.
In this article, we investigate the bijective linear preservers of the sets of
products of two, three, or four involutions in $M_n(\mathbb{F})$.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:25:39 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Li",
"Chi-Kwong",
""
],
[
"Lohan",
"Tejbir",
""
],
[
"Singla",
"Sushil",
""
]
] |
2504.14199 | Jiepeng Fang | Jiepeng Fang, Yixin Lan | Canonical bases of tensor products of integrable highest weight modules
arising from framed constructions | 31pages | null | null | null | math.QA math.RT | http://creativecommons.org/licenses/by/4.0/ | Given a quantum group, we prove that the canonical bases of the tensor
products of its integrable highest weight modules can be obtained from the
canonical bases of the simple integrable highest modules of a bigger quantum
group. As a result, based on the positivity of the canonical bases of the
simple integrable highest modules due to Lusztig, we give an algebraic proof
for the positivity of the canonical bases of the tensor products.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:25:42 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Fang",
"Jiepeng",
""
],
[
"Lan",
"Yixin",
""
]
] |
2504.14200 | Huiyi Chen | Huiyi Chen, Jiawei Peng, Kaihua Tang, Xin Geng, Xu Yang | Enhancing Multimodal In-Context Learning for Image Classification
through Coreset Optimization | 11 pages, 5 figures | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In-context learning (ICL) enables Large Vision-Language Models (LVLMs) to
adapt to new tasks without parameter updates, using a few demonstrations from a
large support set. However, selecting informative demonstrations leads to high
computational and memory costs. While some methods explore selecting a small
and representative coreset in the text classification, evaluating all support
set samples remains costly, and discarded samples lead to unnecessary
information loss. These methods may also be less effective for image
classification due to differences in feature spaces. Given these limitations,
we propose Key-based Coreset Optimization (KeCO), a novel framework that
leverages untapped data to construct a compact and informative coreset. We
introduce visual features as keys within the coreset, which serve as the anchor
for identifying samples to be updated through different selection strategies.
By leveraging untapped samples from the support set, we update the keys of
selected coreset samples, enabling the randomly initialized coreset to evolve
into a more informative coreset under low computational cost. Through extensive
experiments on coarse-grained and fine-grained image classification benchmarks,
we demonstrate that KeCO effectively enhances ICL performance for image
classification task, achieving an average improvement of more than 20\%.
Notably, we evaluate KeCO under a simulated online scenario, and the strong
performance in this scenario highlights the practical value of our framework
for resource-constrained real-world scenarios.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:26:23 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Chen",
"Huiyi",
""
],
[
"Peng",
"Jiawei",
""
],
[
"Tang",
"Kaihua",
""
],
[
"Geng",
"Xin",
""
],
[
"Yang",
"Xu",
""
]
] |
2504.14201 | Jameel-Un Nabi | Jameel-Un Nabi, Tuncay Bayram and Muhammad Majid | Ground-state nuclear properties of neutron-rich copper isotopes and
lepton capture rates in stellar matter | 28 Pages, 12 Figures, 4 Tables | Acta Physica Polonica B, 49(8) (2018) | 10.5506/aphyspolb.49.1531 | null | nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study consists of two separate investigations centered on neutron-rich
isotopes of copper, utilizing two distinct nuclear models. In the first part,
the nuclear ground-state properties of copper isotopes in the mass range 72 <=
A <= 82 were analyzed using the relativistic mean field (RMF) model. Quadrupole
moment-constrained RMF calculations were carried out with DD-ME2 and DD-PC1
density-dependent interactions to compute the ground-state binding energies,
charge radii, proton and neutron radii, quadrupole moments, and deformation
parameters for the 71-82Cu isotopes. The results show good agreement with the
limited experimental data available and previous theoretical predictions. In
addition, potential energy curves were evaluated to investigate the
ground-state geometrical configurations of these isotopes. The second part of
the study is devoted to calculating lepton capture rates under stellar
conditions. While earlier works have provided allowed Gamow-Teller (GT) and
unique first-forbidden (U1F) beta-decay rates for selected neutron-rich Cu
isotopes in stellar environments, the corresponding lepton capture rates had
not yet been computed. This paper presents, for the first time, those lepton
capture rates.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:27:12 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Nabi",
"Jameel-Un",
""
],
[
"Bayram",
"Tuncay",
""
],
[
"Majid",
"Muhammad",
""
]
] |
2504.14202 | Zichuan Liu | Zichuan Liu, Liming Jiang, Qing Yan, Yumin Jia, Hao Kang, Xin Lu | Learning Joint ID-Textual Representation for ID-Preserving Image
Synthesis | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | We propose a novel framework for ID-preserving generation using a multi-modal
encoding strategy rather than injecting identity features via adapters into
pre-trained models. Our method treats identity and text as a unified
conditioning input. To achieve this, we introduce FaceCLIP, a multi-modal
encoder that learns a joint embedding space for both identity and textual
semantics. Given a reference face and a text prompt, FaceCLIP produces a
unified representation that encodes both identity and text, which conditions a
base diffusion model to generate images that are identity-consistent and
text-aligned. We also present a multi-modal alignment algorithm to train
FaceCLIP, using a loss that aligns its joint representation with face, text,
and image embedding spaces. We then build FaceCLIP-SDXL, an ID-preserving image
synthesis pipeline by integrating FaceCLIP with Stable Diffusion XL (SDXL).
Compared to prior methods, FaceCLIP-SDXL enables photorealistic portrait
generation with better identity preservation and textual relevance. Extensive
experiments demonstrate its quantitative and qualitative superiority.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:31:07 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Liu",
"Zichuan",
""
],
[
"Jiang",
"Liming",
""
],
[
"Yan",
"Qing",
""
],
[
"Jia",
"Yumin",
""
],
[
"Kang",
"Hao",
""
],
[
"Lu",
"Xin",
""
]
] |
2504.14203 | Jian Zhang | Jian Zhang, Tianqing Zhang, Qi Li and Hongwei Wang | EIoU-EMC: A Novel Loss for Domain-specific Nested Entity Recognition | Accepted by SIGIR'2025 | null | null | null | cs.CL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, research has mainly focused on the general NER task. There
still have some challenges with nested NER task in the specific domains.
Specifically, the scenarios of low resource and class imbalance impede the wide
application for biomedical and industrial domains. In this study, we design a
novel loss EIoU-EMC, by enhancing the implement of Intersection over Union loss
and Multiclass loss. Our proposed method specially leverages the information of
entity boundary and entity classification, thereby enhancing the model's
capacity to learn from a limited number of data samples. To validate the
performance of this innovative method in enhancing NER task, we conducted
experiments on three distinct biomedical NER datasets and one dataset
constructed by ourselves from industrial complex equipment maintenance
documents. Comparing to strong baselines, our method demonstrates the
competitive performance across all datasets. During the experimental analysis,
our proposed method exhibits significant advancements in entity boundary
recognition and entity classification. Our code are available here.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:31:54 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Jian",
""
],
[
"Zhang",
"Tianqing",
""
],
[
"Li",
"Qi",
""
],
[
"Wang",
"Hongwei",
""
]
] |
2504.14204 | Wenxin Zhang | Wenxin Zhang, Xiaojian Lin, Wenjun Yu, Guangzhen Yao, jingxiang Zhong,
Yu Li, Renda Han, Songcheng Xu, Hao Shi and Cuicui Luo | DConAD: A Differencing-based Contrastive Representation Learning
Framework for Time Series Anomaly Detection | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Time series anomaly detection holds notable importance for risk
identification and fault detection across diverse application domains.
Unsupervised learning methods have become popular because they have no
requirement for labels. However, due to the challenges posed by the
multiplicity of abnormal patterns, the sparsity of anomalies, and the growth of
data scale and complexity, these methods often fail to capture robust and
representative dependencies within the time series for identifying anomalies.
To enhance the ability of models to capture normal patterns of time series and
avoid the retrogression of modeling ability triggered by the dependencies on
high-quality prior knowledge, we propose a differencing-based contrastive
representation learning framework for time series anomaly detection (DConAD).
Specifically, DConAD generates differential data to provide additional
information about time series and utilizes transformer-based architecture to
capture spatiotemporal dependencies, which enhances the robustness of unbiased
representation learning ability. Furthermore, DConAD implements a novel KL
divergence-based contrastive learning paradigm that only uses positive samples
to avoid deviation from reconstruction and deploys the stop-gradient strategy
to compel convergence. Extensive experiments on five public datasets show the
superiority and effectiveness of DConAD compared with nine baselines. The code
is available at https://github.com/shaieesss/DConAD.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:35:06 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Wenxin",
""
],
[
"Lin",
"Xiaojian",
""
],
[
"Yu",
"Wenjun",
""
],
[
"Yao",
"Guangzhen",
""
],
[
"Zhong",
"jingxiang",
""
],
[
"Li",
"Yu",
""
],
[
"Han",
"Renda",
""
],
[
"Xu",
"Songcheng",
""
],
[
"Shi",
"Hao",
""
],
[
"Luo",
"Cuicui",
""
]
] |
2504.14205 | Wenxin Zhang | Wenxin Zhang, Jingxing Zhong, Guangzhen Yao, Renda Han, Xiaojian Lin,
Zeyu Zhang and Cuicui Luo | Dual-channel Heterophilic Message Passing for Graph Fraud Detection | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Fraudulent activities have significantly increased across various domains,
such as e-commerce, online review platforms, and social networks, making fraud
detection a critical task. Spatial Graph Neural Networks (GNNs) have been
successfully applied to fraud detection tasks due to their strong inductive
learning capabilities. However, existing spatial GNN-based methods often
enhance the graph structure by excluding heterophilic neighbors during message
passing to align with the homophilic bias of GNNs. Unfortunately, this approach
can disrupt the original graph topology and increase uncertainty in
predictions. To address these limitations, this paper proposes a novel
framework, Dual-channel Heterophilic Message Passing (DHMP), for fraud
detection. DHMP leverages a heterophily separation module to divide the graph
into homophilic and heterophilic subgraphs, mitigating the low-pass inductive
bias of traditional GNNs. It then applies shared weights to capture signals at
different frequencies independently and incorporates a customized sampling
strategy for training. This allows nodes to adaptively balance the
contributions of various signals based on their labels. Extensive experiments
on three real-world datasets demonstrate that DHMP outperforms existing
methods, highlighting the importance of separating signals with different
frequencies for improved fraud detection. The code is available at
https://github.com/shaieesss/DHMP.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:41:24 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Wenxin",
""
],
[
"Zhong",
"Jingxing",
""
],
[
"Yao",
"Guangzhen",
""
],
[
"Han",
"Renda",
""
],
[
"Lin",
"Xiaojian",
""
],
[
"Zhang",
"Zeyu",
""
],
[
"Luo",
"Cuicui",
""
]
] |
2504.14206 | Wenxin Zhang | Wenxin Zhang and Cuicui Luo | Decomposition-based multi-scale transformer framework for time series
anomaly detection | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Time series anomaly detection is crucial for maintaining stable systems.
Existing methods face two main challenges. First, it is difficult to directly
model the dependencies of diverse and complex patterns within the sequences.
Second, many methods that optimize parameters using mean squared error struggle
with noise in the time series, leading to performance deterioration. To address
these challenges, we propose a transformer-based framework built on
decomposition (TransDe) for multivariate time series anomaly detection. The key
idea is to combine the strengths of time series decomposition and transformers
to effectively learn the complex patterns in normal time series data. A
multi-scale patch-based transformer architecture is proposed to exploit the
representative dependencies of each decomposed component of the time series.
Furthermore, a contrastive learn paradigm based on patch operation is proposed,
which leverages KL divergence to align the positive pairs, namely the pure
representations of normal patterns between different patch-level views. A novel
asynchronous loss function with a stop-gradient strategy is further introduced
to enhance the performance of TransDe effectively. It can avoid time-consuming
and labor-intensive computation costs in the optimization process. Extensive
experiments on five public datasets are conducted and TransDe shows superiority
compared with twelve baselines in terms of F1 score. Our code is available at
https://github.com/shaieesss/TransDe.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:47:38 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Wenxin",
""
],
[
"Luo",
"Cuicui",
""
]
] |
2504.14207 | Hirofumi Wada | Takuro Kataoka, Taiju Yoneda, Daisuke Nakane and Hirofumi Wada | Bacteria exploit torque-induced buckling instability for flagellar
wrapping | 9 pages, 6 figures, plus SI 9 pages, 7 figures | null | null | null | cond-mat.soft physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | Recent advances in microscopy techniques has uncovered unique aspects of
flagella-driven motility in bacteria. A remarkable example is the discovery of
flagellar wrapping, a phenomenon whereby a bacterium wraps its flagellum (or
flagellar bundle) around its cell body and propels itself like a corkscrew,
enabling locomotion in highly viscous or confined environments. For certain
bacterial species, this flagellar-wrapping mode is crucial for establishing
selective symbiotic relationships with their hosts. The transformation of a
flagellum from an extended to a folded (wrapped) state is triggered by a
buckling instability driven by the motor-generated torque that unwinds the
helical filament. This study investigated this biologically inspired, novel
buckling mechanism through a combination of macroscale physical experiments,
numerical simulations, and scaling theory to reveal its underlying physical
principles. Excellent quantitative agreement between experiments and numerical
results showed that long-range hydrodynamic interactions (HIs) are essential
for accurate quantitative descriptions of the geometrically nonlinear
deformation of the helical filament during wrapping. By systematically
analyzing extensive experimental and numerical data, we constructed a stability
diagram that rationalized the stability boundary through an elastohydrodynamic
scaling analysis. Leveraging the scaling nature of this study, we compared our
physical results with available biological data and demonstrated that bacteria
exploit motor-induced buckling instability to initiate their flagellar
wrapping. Our findings indicate that this mechanically-driven process is
essential to bacterial-wrapping motility and consequently, plays a critical
role in symbiosis and infection.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:51:43 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Kataoka",
"Takuro",
""
],
[
"Yoneda",
"Taiju",
""
],
[
"Nakane",
"Daisuke",
""
],
[
"Wada",
"Hirofumi",
""
]
] |
2504.14208 | Mingzhe Han | Mingzhe Han, Dongsheng Li, Jiafeng Xia, Jiahao Liu, Hansu Gu, Peng
Zhang, Ning Gu and Tun Lu | FedCIA: Federated Collaborative Information Aggregation for
Privacy-Preserving Recommendation | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recommendation algorithms rely on user historical interactions to deliver
personalized suggestions, which raises significant privacy concerns. Federated
recommendation algorithms tackle this issue by combining local model training
with server-side model aggregation, where most existing algorithms use a
uniform weighted summation to aggregate item embeddings from different client
models. This approach has three major limitations: 1) information loss during
aggregation, 2) failure to retain personalized local features, and 3)
incompatibility with parameter-free recommendation algorithms. To address these
limitations, we first review the development of recommendation algorithms and
recognize that their core function is to share collaborative information,
specifically the global relationship between users and items. With this
understanding, we propose a novel aggregation paradigm named collaborative
information aggregation, which focuses on sharing collaborative information
rather than item parameters. Based on this new paradigm, we introduce the
federated collaborative information aggregation (FedCIA) method for
privacy-preserving recommendation. This method requires each client to upload
item similarity matrices for aggregation, which allows clients to align their
local models without constraining embeddings to a unified vector space. As a
result, it mitigates information loss caused by direct summation, preserves the
personalized embedding distributions of individual clients, and supports the
aggregation of parameter-free models. Theoretical analysis and experimental
results on real-world datasets demonstrate the superior performance of FedCIA
compared with the state-of-the-art federated recommendation algorithms. Code is
available at https://github.com/Mingzhe-Han/FedCIA.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 06:59:34 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Han",
"Mingzhe",
""
],
[
"Li",
"Dongsheng",
""
],
[
"Xia",
"Jiafeng",
""
],
[
"Liu",
"Jiahao",
""
],
[
"Gu",
"Hansu",
""
],
[
"Zhang",
"Peng",
""
],
[
"Gu",
"Ning",
""
],
[
"Lu",
"Tun",
""
]
] |
2504.14209 | Xiangkai Ma | Xiangkai Ma, Xiaobin Hong, Wenzhong Li, Sanglu Lu | Pets: General Pattern Assisted Architecture For Time Series Analysis | null | null | null | null | cs.AI | http://creativecommons.org/publicdomain/zero/1.0/ | Time series analysis has found widespread applications in areas such as
weather forecasting, anomaly detection, and healthcare. However, real-world
sequential data often exhibit a superimposed state of various fluctuation
patterns, including hourly, daily, and monthly frequencies. Traditional
decomposition techniques struggle to effectively disentangle these multiple
fluctuation patterns from the seasonal components, making time series analysis
challenging. Surpassing the existing multi-period decoupling paradigms, this
paper introduces a novel perspective based on energy distribution within the
temporal-spectrum space. By adaptively quantifying observed sequences into
continuous frequency band intervals, the proposed approach reconstructs
fluctuation patterns across diverse periods without relying on domain-specific
prior knowledge. Building upon this innovative strategy, we propose Pets, an
enhanced architecture that is adaptable to arbitrary model structures. Pets
integrates a Fluctuation Pattern Assisted (FPA) module and a Context-Guided
Mixture of Predictors (MoP). The FPA module facilitates information fusion
among diverse fluctuation patterns by capturing their dependencies and
progressively modeling these patterns as latent representations at each layer.
Meanwhile, the MoP module leverages these compound pattern representations to
guide and regulate the reconstruction of distinct fluctuations hierarchically.
Pets achieves state-of-the-art performance across various tasks, including
forecasting, imputation, anomaly detection, and classification, while
demonstrating strong generalization and robustness.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:12:57 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ma",
"Xiangkai",
""
],
[
"Hong",
"Xiaobin",
""
],
[
"Li",
"Wenzhong",
""
],
[
"Lu",
"Sanglu",
""
]
] |
2504.14210 | Kedar Khare | N Goyal and K Khare | Phase tomography with axial structured illumination | null | null | null | null | eess.IV physics.optics | http://creativecommons.org/licenses/by/4.0/ | Holographic Tomography (HT) or Optical Diffraction Tomography (ODT) provides
slice-by-slice information about the refractive index (RI) of three-dimensional
(3D) samples and is emerging as an important label-free imaging modality for
Life sciences. HT systems go beyond the digital holographic microscopy (DHM)
systems that provide a two-dimensional (2D) representation of the total
accumulated phase acquired by a plane beam on transmission through a 3D sample.
While the early HT systems used a direct reconstruction methodology based on
the Fourier diffraction theorem, in recent years, there has been an increasing
shift towards using iterative optimization frameworks for solving the 3D RI
reconstruction problem. Despite this algorithmic framework shift, the HT system
hardware still largely uses the multi-angle illumination geometries that were
suitable for reconstructions based on the Fourier diffraction theorem. The
present work examines the possibility of HT reconstruction through the use of
on-axis structured illumination(s) that nominally illuminates the 3D sample
along the axial direction. Through a simulation study, it is shown that a
cross-talk free slice-by-slice 3D RI reconstruction of the sample is possible
in this case via the use of sparsity penalties if the slice-to-slice distance
obeys a design curve based on the notion of effective depth of focus. The
simulation results for two-, three- and four-slice 3D objects with laterally
overlapping features clearly outline the separate roles played by the
slice-to-slice de-correlation of the field propagating through the 3D sample
and that of the sparsity penalty used to guide the iterative solution. Our
results suggest the possibility of realizing an Axial Structured Illumination
Tomography (ASIT) system configuration that avoids the use of
hardware-intensive multi-angle illumination geometry.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:13:58 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Goyal",
"N",
""
],
[
"Khare",
"K",
""
]
] |
2504.14211 | Xiaojun Zhou | Xiaojun Zhou, Chunhua Yang, Weihua Gui | Efficient state transition algorithm with guaranteed optimality | 13 pages | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a constructivism-based intelligent optimization method, state transition
algorithm (STA) has exhibited powerful search ability in optimization. However,
the standard STA still shows slow convergence at a later stage for flat
landscape and a user has to preset its maximum number of iterations (or
function evaluations) by experience. To resolve these two issues, efficient
state transition algorithm is proposed with guaranteed optimality. Firstly,
novel translation transformations based on predictive modeling are proposed to
generate more potential candidates by utilizing historical information.
Secondly, parameter control strategies are proposed to accelerate the
convergence. Thirdly, a specific termination condition is designed to guarantee
that the STA can stop automatically at an optimal point, which is equivalent to
the zero gradient in mathematical programming. Experimental results have
demonstrated the effectiveness and superiority of the proposed method.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:23:18 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhou",
"Xiaojun",
""
],
[
"Yang",
"Chunhua",
""
],
[
"Gui",
"Weihua",
""
]
] |
2504.14212 | Takuma Udagawa | Takuma Udagawa, Yang Zhao, Hiroshi Kanayama, Bishwaranjan
Bhattacharjee | Bias Analysis and Mitigation through Protected Attribute Detection and
Regard Classification | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) acquire general linguistic knowledge from
massive-scale pretraining. However, pretraining data mainly comprised of
web-crawled texts contain undesirable social biases which can be perpetuated or
even amplified by LLMs. In this study, we propose an efficient yet effective
annotation pipeline to investigate social biases in the pretraining corpora.
Our pipeline consists of protected attribute detection to identify diverse
demographics, followed by regard classification to analyze the language
polarity towards each attribute. Through our experiments, we demonstrate the
effect of our bias analysis and mitigation measures, focusing on Common Crawl
as the most representative pretraining corpus.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:36:02 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Udagawa",
"Takuma",
""
],
[
"Zhao",
"Yang",
""
],
[
"Kanayama",
"Hiroshi",
""
],
[
"Bhattacharjee",
"Bishwaranjan",
""
]
] |
2504.14213 | Evgeniy Petrov | Ravindra K. Bisht and Evgeniy Petrov | Fixed point results for multipoint Kannan-type mappings | 12 pages | null | null | null | math.GN | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce and study a new type of mappings in metric spaces termed
$n$-point Kannan-type mappings. A fixed-point theorem is proved for these
mappings. In general case such mappings are discontinuous in the domain but
necessarily continuous at fixed points. Conditions under which usual Kannan
mappings and mapping contracting the total pairwise distances between $n$
points are $n$-point Kannan-type mappings are found. It is shown that
additional conditions of asymptotic regularity and continuity allow to extend
the value of the contraction coefficient in fixed-point theorems for $n$-point
Kannan-type mappings.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:36:11 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Bisht",
"Ravindra K.",
""
],
[
"Petrov",
"Evgeniy",
""
]
] |
2504.14214 | Hongji Li | Hongji Li, Hanwen Du, Youhua Li, Junchen Fu, Chunxiao Li, Ziyi Zhuang,
Jiakang Li, and Yongxin Ni | Teach Me How to Denoise: A Universal Framework for Denoising Multi-modal
Recommender Systems via Guided Calibration | Accepted to ACM Web Search and Data Mining (WSDM) 2025 | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The surge in multimedia content has led to the development of Multi-Modal
Recommender Systems (MMRecs), which use diverse modalities such as text,
images, videos, and audio for more personalized recommendations. However,
MMRecs struggle with noisy data caused by misalignment among modal content and
the gap between modal semantics and recommendation semantics. Traditional
denoising methods are inadequate due to the complexity of multi-modal data. To
address this, we propose a universal guided in-sync distillation denoising
framework for multi-modal recommendation (GUIDER), designed to improve MMRecs
by denoising user feedback. Specifically, GUIDER uses a re-calibration strategy
to identify clean and noisy interactions from modal content. It incorporates a
Denoising Bayesian Personalized Ranking (DBPR) loss function to handle implicit
user feedback. Finally, it applies a denoising knowledge distillation objective
based on Optimal Transport distance to guide the alignment from modality
representations to recommendation semantics. GUIDER can be seamlessly
integrated into existing MMRecs methods as a plug-and-play solution.
Experimental results on four public datasets demonstrate its effectiveness and
generalizability. Our source code is available at
https://github.com/Neon-Jing/Guider
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:37:03 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Li",
"Hongji",
""
],
[
"Du",
"Hanwen",
""
],
[
"Li",
"Youhua",
""
],
[
"Fu",
"Junchen",
""
],
[
"Li",
"Chunxiao",
""
],
[
"Zhuang",
"Ziyi",
""
],
[
"Li",
"Jiakang",
""
],
[
"Ni",
"Yongxin",
""
]
] |
2504.14216 | Pierre-Alain Fayolle | Pierre-Alain Fayolle and Evgenii Maltsev | PyFRep: Shape Modeling with Differentiable Function Representation | 21 pages, 10 figures. Code available at
https://github.com/fayolle/PyFRep | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a framework for performing differentiable geometric modeling based
on the Function Representation (FRep). The framework is built on top of modern
libraries for performing automatic differentiation allowing us to obtain
derivatives w.r.t. space or shape parameters. We demonstrate possible
applications of this framework: Curvature estimation for shape interrogation,
signed distance function computation and approximation and fitting shape
parameters of a parametric model to data. Our framework is released as
open-source.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:43:34 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Fayolle",
"Pierre-Alain",
""
],
[
"Maltsev",
"Evgenii",
""
]
] |
2504.14217 | Frances Allen | Dana O. Byrne, Stephanie M. Ribet, Demie Kepaptsoglou, Quentin M.
Ramasse, Colin Ophus, Frances I. Allen | Fabrication and characterization of boron-terminated tetravacancies in
monolayer hBN using STEM, EELS and electron ptychography | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Tetravacancies in monolayer hexagonal boron nitride (hBN) with consistent
edge termination (boron or nitrogen) form triangular nanopores with
electrostatic potentials that can be leveraged for applications such as
selective ion transport and neuromorphic computing. In order to quantitatively
predict the properties of these structures, an atomic-level understanding of
their local electronic and chemical environments is required. Moreover, robust
methods for their precision manufacture are needed. Here were use electron
irradiation in a scanning transmission electron microscope (STEM) at high dose
rate to drive the formation of boron-terminated tetravacancies in monolayer
hBN. Characterization of the defects is achieved using aberration-corrected
STEM, monochromated electron energy-loss spectroscopy (EELS), and electron
ptychography. Z-contrast in STEM and chemical fingerprinting by core-loss EELS
enable identification of the edge terminations, while electron ptychography
gives insight into structural relaxation of the tetravacancies and provides
evidence of enhanced electron density around the defect perimeters indicative
of bonding effects.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:48:59 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Byrne",
"Dana O.",
""
],
[
"Ribet",
"Stephanie M.",
""
],
[
"Kepaptsoglou",
"Demie",
""
],
[
"Ramasse",
"Quentin M.",
""
],
[
"Ophus",
"Colin",
""
],
[
"Allen",
"Frances I.",
""
]
] |
2504.14218 | Junchi Yao | Junchi Yao, Shu Yang, Jianhua Xu, Lijie Hu, Mengdi Li, Di Wang | Understanding the Repeat Curse in Large Language Models from a Feature
Perspective | Submitted to ACL 2025 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) have made remarkable progress in various
domains, yet they often suffer from repetitive text generation, a phenomenon we
refer to as the "Repeat Curse". While previous studies have proposed decoding
strategies to mitigate repetition, the underlying mechanism behind this issue
remains insufficiently explored. In this work, we investigate the root causes
of repetition in LLMs through the lens of mechanistic interpretability.
Inspired by recent advances in Sparse Autoencoders (SAEs), which enable
monosemantic feature extraction, we propose a novel approach, "Duplicatus
Charm", to induce and analyze the Repeat Curse. Our method systematically
identifies "Repetition Features" -the key model activations responsible for
generating repetitive outputs. First, we locate the layers most involved in
repetition through logit analysis. Next, we extract and stimulate relevant
features using SAE-based activation manipulation. To validate our approach, we
construct a repetition dataset covering token and paragraph level repetitions
and introduce an evaluation pipeline to quantify the influence of identified
repetition features. Furthermore, by deactivating these features, we have
effectively mitigated the Repeat Curse.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 07:53:37 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Yao",
"Junchi",
""
],
[
"Yang",
"Shu",
""
],
[
"Xu",
"Jianhua",
""
],
[
"Hu",
"Lijie",
""
],
[
"Li",
"Mengdi",
""
],
[
"Wang",
"Di",
""
]
] |
2504.14219 | Alara Dirik | Alara Dirik, Tuanfeng Wang, Duygu Ceylan, Stefanos Zafeiriou, Anna
Fr\"uhst\"uck | PRISM: A Unified Framework for Photorealistic Reconstruction and
Intrinsic Scene Modeling | null | null | null | null | cs.GR cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present PRISM, a unified framework that enables multiple image generation
and editing tasks in a single foundational model. Starting from a pre-trained
text-to-image diffusion model, PRISM proposes an effective fine-tuning strategy
to produce RGB images along with intrinsic maps (referred to as X layers)
simultaneously. Unlike previous approaches, which infer intrinsic properties
individually or require separate models for decomposition and conditional
generation, PRISM maintains consistency across modalities by generating all
intrinsic layers jointly. It supports diverse tasks, including text-to-RGBX
generation, RGB-to-X decomposition, and X-to-RGBX conditional generation.
Additionally, PRISM enables both global and local image editing through
conditioning on selected intrinsic layers and text prompts. Extensive
experiments demonstrate the competitive performance of PRISM both for intrinsic
image decomposition and conditional image generation while preserving the base
model's text-to-image generation capability.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:03:06 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Dirik",
"Alara",
""
],
[
"Wang",
"Tuanfeng",
""
],
[
"Ceylan",
"Duygu",
""
],
[
"Zafeiriou",
"Stefanos",
""
],
[
"Frühstück",
"Anna",
""
]
] |
2504.14220 | Jukka Ruohonen | Jukka Ruohonen and Kalle Rindell and Simone Busetti | From Cyber Security Incident Management to Cyber Security Crisis
Management in the European Union | Submitted to COSE (VSI: Cybersecurity, Privacy, Trust and Regulation) | null | null | null | cs.CR cs.CY | http://creativecommons.org/licenses/by/4.0/ | Incident management is a classical topic in cyber security. Recently, the
European Union (EU) has started to consider also the relation between cyber
security incidents and cyber security crises. These considerations and
preparations, including those specified in the EU's new cyber security laws,
constitute the paper's topic. According to an analysis of the laws and
associated policy documents, (i) cyber security crises are equated in the EU to
large-scale cyber security incidents that either exceed a handling capacity of
a single member state or affect at least two member states. For this and other
purposes, (ii) the new laws substantially increase mandatory reporting about
cyber security incidents, including but not limited to the large-scale
incidents. Despite the laws and new governance bodies established by them,
however, (iii) the working of actual cyber security crisis management remains
unclear particularly at the EU-level. With these policy research results, the
paper advances the domain of cyber security incident management research by
elaborating how European law perceives cyber security crises and their relation
to cyber security incidents, paving the way for many relevant further research
topics with practical relevance, whether theoretical, conceptual, or empirical.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:03:29 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ruohonen",
"Jukka",
""
],
[
"Rindell",
"Kalle",
""
],
[
"Busetti",
"Simone",
""
]
] |
2504.14221 | Bo Peng | Wenbing Zhu, Lidong Wang, Ziqing Zhou, Chengjie Wang, Yurui Pan, Ruoyi
Zhang, Zhuhao Chen, Linjie Cheng, Bin-Bin Gao, Jiangning Zhang, Zhenye Gan,
Yuxie Wang, Yulong Chen, Shuguang Qian, Mingmin Chi, Bo Peng, Lizhuang Ma | Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly
Detection | 13 pages. Dataset and code: https://realiad4ad.github.io/Real-IAD D3 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The increasing complexity of industrial anomaly detection (IAD) has
positioned multimodal detection methods as a focal area of machine vision
research. However, dedicated multimodal datasets specifically tailored for IAD
remain limited. Pioneering datasets like MVTec 3D have laid essential
groundwork in multimodal IAD by incorporating RGB+3D data, but still face
challenges in bridging the gap with real industrial environments due to
limitations in scale and resolution. To address these challenges, we introduce
Real-IAD D3, a high-precision multimodal dataset that uniquely incorporates an
additional pseudo3D modality generated through photometric stereo, alongside
high-resolution RGB images and micrometer-level 3D point clouds. Real-IAD D3
features finer defects, diverse anomalies, and greater scale across 20
categories, providing a challenging benchmark for multimodal IAD Additionally,
we introduce an effective approach that integrates RGB, point cloud, and
pseudo-3D depth information to leverage the complementary strengths of each
modality, enhancing detection performance. Our experiments highlight the
importance of these modalities in boosting detection robustness and overall IAD
performance. The dataset and code are publicly accessible for research purposes
at https://realiad4ad.github.io/Real-IAD D3
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:05:47 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhu",
"Wenbing",
""
],
[
"Wang",
"Lidong",
""
],
[
"Zhou",
"Ziqing",
""
],
[
"Wang",
"Chengjie",
""
],
[
"Pan",
"Yurui",
""
],
[
"Zhang",
"Ruoyi",
""
],
[
"Chen",
"Zhuhao",
""
],
[
"Cheng",
"Linjie",
""
],
[
"Gao",
"Bin-Bin",
""
],
[
"Zhang",
"Jiangning",
""
],
[
"Gan",
"Zhenye",
""
],
[
"Wang",
"Yuxie",
""
],
[
"Chen",
"Yulong",
""
],
[
"Qian",
"Shuguang",
""
],
[
"Chi",
"Mingmin",
""
],
[
"Peng",
"Bo",
""
],
[
"Ma",
"Lizhuang",
""
]
] |
2504.14222 | Diego Gomez-Zara | Mohammed Almutairi, Charles Chiang, Yuxin Bai, Diego Gomez-Zara | tAIfa: Enhancing Team Effectiveness and Cohesion with AI-Generated
Automated Feedback | 25 pages, CHIWORK '25: Proceedings of the 4th Annual Symposium on
Human-Computer Interaction for Work | null | 10.1145/3729176.3729197 | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | Providing timely and actionable feedback is crucial for effective
collaboration, learning, and coordination within teams. However, many teams
face challenges in receiving feedback that aligns with their goals and promotes
cohesion. We introduce tAIfa (``Team AI Feedback Assistant''), an AI agent that
uses Large Language Models (LLMs) to provide personalized, automated feedback
to teams and their members. tAIfa analyzes team interactions, identifies
strengths and areas for improvement, and delivers targeted feedback based on
communication patterns. We conducted a between-subjects study with 18 teams
testing whether using tAIfa impacted their teamwork. Our findings show that
tAIfa improved communication and contributions within the teams. This paper
contributes to the Human-AI Interaction literature by presenting a
computational framework that integrates LLMs to provide automated feedback,
introducing tAIfa as a tool to enhance team engagement and cohesion, and
providing insights into future AI applications to support team collaboration.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:06:48 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Almutairi",
"Mohammed",
""
],
[
"Chiang",
"Charles",
""
],
[
"Bai",
"Yuxin",
""
],
[
"Gomez-Zara",
"Diego",
""
]
] |
2504.14223 | Michael F\"arber | Michael F\"arber, Parisa Aghdam, Kyuri Im, Mario Tawfelis, Hardik
Ghoshal | SimplifyMyText: An LLM-Based System for Inclusive Plain Language Text
Simplification | accepted at ECIR 2025 | null | null | null | cs.CL cs.AI cs.HC | http://creativecommons.org/licenses/by/4.0/ | Text simplification is essential for making complex content accessible to
diverse audiences who face comprehension challenges. Yet, the limited
availability of simplified materials creates significant barriers to personal
and professional growth and hinders social inclusion. Although researchers have
explored various methods for automatic text simplification, none fully leverage
large language models (LLMs) to offer tailored customization for different
target groups and varying levels of simplicity. Moreover, despite its proven
benefits for both consumers and organizations, the well-established practice of
plain language remains underutilized. In this paper, we
https://simplifymytext.org, the first system designed to produce plain language
content from multiple input formats, including typed text and file uploads,
with flexible customization options for diverse audiences. We employ GPT-4 and
Llama-3 and evaluate outputs across multiple metrics. Overall, our work
contributes to research on automatic text simplification and highlights the
importance of tailored communication in promoting inclusivity.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:07:53 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Färber",
"Michael",
""
],
[
"Aghdam",
"Parisa",
""
],
[
"Im",
"Kyuri",
""
],
[
"Tawfelis",
"Mario",
""
],
[
"Ghoshal",
"Hardik",
""
]
] |
2504.14224 | Yongguang Li | Yongguang Li, Jindong Li, Qi Wang, Qianli Xing, Runliang Niu,
Shengsheng Wang and Menglin Yang | Revisiting CLIP for SF-OSDA: Unleashing Zero-Shot Potential with
Adaptive Threshold and Training-Free Feature Filtering | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Source-Free Unsupervised Open-Set Domain Adaptation (SF-OSDA) methods using
CLIP face significant issues: (1) while heavily dependent on domain-specific
threshold selection, existing methods employ simple fixed thresholds,
underutilizing CLIP's zero-shot potential in SF-OSDA scenarios; and (2)
overlook intrinsic class tendencies while employing complex training to enforce
feature separation, incurring deployment costs and feature shifts that
compromise CLIP's generalization ability. To address these issues, we propose
CLIPXpert, a novel SF-OSDA approach that integrates two key components: an
adaptive thresholding strategy and an unknown class feature filtering module.
Specifically, the Box-Cox GMM-Based Adaptive Thresholding (BGAT) module
dynamically determines the optimal threshold by estimating sample score
distributions, balancing known class recognition and unknown class sample
detection. Additionally, the Singular Value Decomposition (SVD)-Based
Unknown-Class Feature Filtering (SUFF) module reduces the tendency of unknown
class samples towards known classes, improving the separation between known and
unknown classes. Experiments show that our source-free and training-free method
outperforms state-of-the-art trained approach UOTA by 1.92% on the DomainNet
dataset, achieves SOTA-comparable performance on datasets such as Office-Home,
and surpasses other SF-OSDA methods. This not only validates the effectiveness
of our proposed method but also highlights CLIP's strong zero-shot potential
for SF-OSDA tasks.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:12:19 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Li",
"Yongguang",
""
],
[
"Li",
"Jindong",
""
],
[
"Wang",
"Qi",
""
],
[
"Xing",
"Qianli",
""
],
[
"Niu",
"Runliang",
""
],
[
"Wang",
"Shengsheng",
""
],
[
"Yang",
"Menglin",
""
]
] |
2504.14225 | Sihao Chen | Bowen Jiang, Zhuoqun Hao, Young-Min Cho, Bryan Li, Yuan Yuan, Sihao
Chen, Lyle Ungar, Camillo J. Taylor, Dan Roth | Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and
Personalized Responses at Scale | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Large Language Models (LLMs) have emerged as personalized assistants for
users across a wide range of tasks -- from offering writing support to
delivering tailored recommendations or consultations. Over time, the
interaction history between a user and an LLM can provide extensive information
about an individual's traits and preferences. However, open questions remain on
how well LLMs today can effectively leverage such history to (1) internalize
the user's inherent traits and preferences, (2) track how the user profiling
and preferences evolve over time, and (3) generate personalized responses
accordingly in new scenarios.
In this work, we introduce the PERSONAMEM benchmark. PERSONAMEM features
curated user profiles with over 180 simulated user-LLM interaction histories,
each containing up to 60 sessions of multi-turn conversations across 15
real-world tasks that require personalization. Given an in-situ user query,
i.e. query issued by the user from the first-person perspective, we evaluate
LLM chatbots' ability to identify the most suitable response according to the
current state of the user's profile. We observe that current LLMs still
struggle to recognize the dynamic evolution in users' profiles over time
through direct prompting approaches. As a consequence, LLMs often fail to
deliver responses that align with users' current situations and preferences,
with frontier models such as GPT-4.1, o4-mini, GPT-4.5, o1, or Gemini-2.0
achieving only around 50% overall accuracy, suggesting room for improvement. We
hope that PERSONAMEM, along with the user profile and conversation simulation
pipeline, can facilitate future research in the development of truly user-aware
chatbots. Code and data are available at github.com/bowen-upenn/PersonaMem.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:16:10 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Jiang",
"Bowen",
""
],
[
"Hao",
"Zhuoqun",
""
],
[
"Cho",
"Young-Min",
""
],
[
"Li",
"Bryan",
""
],
[
"Yuan",
"Yuan",
""
],
[
"Chen",
"Sihao",
""
],
[
"Ungar",
"Lyle",
""
],
[
"Taylor",
"Camillo J.",
""
],
[
"Roth",
"Dan",
""
]
] |
2504.14226 | Chandrashekhar Rai | Chandrashekhar Rai and Debarati Sen | An Artificial Intelligence Enabled Signature Estimation of Dual Wideband
Systems in Ultra-Low Signal-to-Noise Ratio | 13 pages, 19 figures, 4tables | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Millimeter-wave (mmWave) massive Multiple Input Multiple Output (MIMO)
systems encounter both spatial wideband spreading and temporal wideband effects
in the communication channels of individual users. Accurate estimation of a
user's channel signature -- specifically, the direction of arrival and time of
arrival -- is crucial for designing efficient beamforming transceivers,
especially under noisy observations. In this work, we propose an Artificial
Intelligence (AI)-enabled framework for estimating the channel signature of a
user's location in mmWave massive MIMO systems. Our approach explicitly
accounts for spatial wideband spreading, finite basis leakage effects, and
significant unknown receiver noise. We demonstrate the effectiveness of a
denoising convolutional neural network with residual learning for recovering
channel responses, even when channel gains are of extremely low amplitude and
embedded in ultra-high receiver noise environments. Notably, our method
successfully recovers spatio-temporal diversity branches at signal-to-noise
ratios as low as -20 dB. Furthermore, we introduce a local gravitation-based
clustering algorithm to infer the number of physical propagation paths (unknown
a priori) and to identify their respective support in the delay-angle domain of
the denoised response. To complement our approach, we design tailored metrics
for evaluating denoising and clustering performance within the context of
wireless communications. We validate our framework through system-level
simulations using Orthogonal Frequency Division Multiplexing (OFDM) with a
Quadrature Phase Shift Keying (QPSK) modulation scheme over mmWave fading
channels, highlighting the necessity and robustness of the proposed methods in
ultra-low SNR scenarios.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:20:16 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Rai",
"Chandrashekhar",
""
],
[
"Sen",
"Debarati",
""
]
] |
2504.14227 | Ivar Bengtsson M.Sc. | Ivar Bengtsson | Reducing the set of considered scenarios in robust optimization of
intensity-modulated proton therapy | null | null | null | null | physics.med-ph math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Robust optimization is a commonly employed method to mitigate uncertainties
in the planning of intensity-modulated proton therapy (IMPT). In certain
contexts, the large number of uncertainty scenarios makes the robust problem
impractically expensive to solve. Recent developments in research on IMPT
treatment planning indicate that the number of ideally considered error
scenarios may continue to increase.
In this paper, we therefore investigate methods that reduce the size of the
scenario set considered during the robust optimization. Six cases of patients
with non-small cell lung cancer are considered. First, we investigate the
existence of an optimal subset of scenarios that needs to be considered during
robust optimization, and perform experiments to see if the set can be found in
a reasonable time and substitute for the full set of scenarios during robust
IMPT optimization. We then consider heuristic methods to estimate this subset
or find subsets with similar properties. Specifically, we select a subset of
maximal diversity in terms of scenario-specific features such as the dose
distributions and function gradients at the initial point. Finally, we consider
adversarial methods as an alternative to solving the full robust problem and
investigate the impact on computation times.
The results indicate that the optimal subset can be well approximated by
solving the robust IMPT problem with conventional methods. Of the methods
designed to approximate it within a practically useful time frame, the results
of the diversity-maximization methods indicate that they may perform better
than a manual selection of scenarios based on the patient geometry. In
addition, the adversarial approaches decreased the computation time by at least
half compared to the conventional approach.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:25:36 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Bengtsson",
"Ivar",
""
]
] |
2504.14228 | Mihkel Kama | Mihkel Kama, Oliver Shorttle, Sandipan P. D. Borthakur, Luke Keyte,
Jennifer B. Bergner, Luca Fossati, Colin P. Folsom, Heleri Ramler | Refractory phosphorus in the HD 100546 protoplanetary disk | 11 pages, 7 figures, accepted for publication in MNRAS | null | null | null | astro-ph.EP astro-ph.SR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The phosphorus budget of planets is intertwined with their formation history
and is thought to influence their habitability. The chemical reservoirs and
volatile \emph{vs} refractory budget of phosphorus in planet-forming
environments have so far eluded empirical characterisation. We employ
high-resolution spectra from HST/STIS in the ultraviolet and APEX in the sub-mm
to constrain the phosphorus budget in the well-characterized HD\,100546 star
and protoplanetary disk system. We measure
$\log{(P/H)_{\star}}=-7.50^{+0.23}_{-0.28}$ on the stellar surface, which
traces the total inventory of P in accreting gas \emph{and }dust from the inner
disk. The inner disk gas, inside of the main dust trap, has $\log{(P/H)_{\rm
in}}\lesssim-8.70$, and the outer disk gas $\log{(P/H)_{\rm
out}}\lesssim-9.30$. Phosphorus in the disk is carried by a relatively
refractory reservoir, consistent with minerals such as apatite or
schreibersite, or with ammonium phosphate salts, in terms of sublimation
temperature. We discuss the impact this might have on the two protoplanets
around HD\,100546. Our results contribute to our understanding of the chemical
habitability of planetary systems and lay a foundation for future explorations,
especially in the context of JWST and \emph{Ariel} which can study phosphorus
in exoplanet atmospheres.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:35:00 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Kama",
"Mihkel",
""
],
[
"Shorttle",
"Oliver",
""
],
[
"Borthakur",
"Sandipan P. D.",
""
],
[
"Keyte",
"Luke",
""
],
[
"Bergner",
"Jennifer B.",
""
],
[
"Fossati",
"Luca",
""
],
[
"Folsom",
"Colin P.",
""
],
[
"Ramler",
"Heleri",
""
]
] |
2504.14229 | Laszlo Arpad Gergely | Bence Juh\'asz, L\'aszl\'o \'Arp\'ad Gergely | K-essence sources of Kerr-Schild spacetimes | 20 pages | Universe 11, 100 (2025) | 10.3390/universe11030100 | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We extend a result by one of the authors, established for nonvacuum Einstein
gravity, to minimally coupled k-essence scalar-tensor theories. First we prove
that in order to source a Kerr-Schild type spacetime, the k-essence Lagrangian
should be at most quadratic in the kinetic term. This is reduced to linear
dependence when the Kerr-Schild null congruence is autoparallel. Finally, we
show that requiring the solutions of the Einstein equations linearized in
Kerr-Schild type perturbations to also solve the full nonlinear system of
Einstein equations, selects once again k-essence scalar fields with Lagrangians
linear in the kinetic term. The only other k-essence sharing the property of
sourcing perturbative Kerr-Schild spacetimes which are also exact, is the
scalar field constant along the integral curves of the Kerr-Schild congruence,
with otherwise unrestricted Lagrangian.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:38:48 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Juhász",
"Bence",
""
],
[
"Gergely",
"László Árpád",
""
]
] |
2504.14230 | Erfang Shan | Songtao He, Erfang Shan, Hanqi Zhou | Highly mutually dependent unions and new axiomatizations of the Owen
value | 20 | null | null | null | econ.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Owen value is an well-known allocation rule for cooperative games with
coalition structure.In this paper, we introduce the concept of highly mutually
dependent unions. Two unions in a cooperative game with coalition structure are
said to be highly mutually dependent if any pair of players, with one from each
of the two unions, are mutually dependent in the game.Based on this concept, we
introduce two axioms: weak mutually dependent between unions and differential
marginality of inter-mutually dependent unions. Furthermore, we also propose
another two axioms: super inter-unions marginality and invariance across games,
where the former one is based on the concept of the inter-unions marginal
contribution. By using the axioms and combining with some standard axioms, we
present three axiomatic characterizations of the Owen value.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:44:42 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"He",
"Songtao",
""
],
[
"Shan",
"Erfang",
""
],
[
"Zhou",
"Hanqi",
""
]
] |
2504.14231 | Johannes Spoecklberger | Johannes Spoecklberger, Wei Lin, Pedro Hermosilla, Sivan Doveh, Horst
Possegger, M. Jehanzeb Mirza | Exploring Modality Guidance to Enhance VFM-based Feature Fusion for UDA
in 3D Semantic Segmentation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Vision Foundation Models (VFMs) have become a de facto choice for many
downstream vision tasks, like image classification, image segmentation, and
object localization. However, they can also provide significant utility for
downstream 3D tasks that can leverage the cross-modal information (e.g., from
paired image data). In our work, we further explore the utility of VFMs for
adapting from a labeled source to unlabeled target data for the task of
LiDAR-based 3D semantic segmentation. Our method consumes paired 2D-3D (image
and point cloud) data and relies on the robust (cross-domain) features from a
VFM to train a 3D backbone on a mix of labeled source and unlabeled target
data. At the heart of our method lies a fusion network that is guided by both
the image and point cloud streams, with their relative contributions adjusted
based on the target domain. We extensively compare our proposed methodology
with different state-of-the-art methods in several settings and achieve strong
performance gains. For example, achieving an average improvement of 6.5 mIoU
(over all tasks), when compared with the previous state-of-the-art.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:53:54 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Spoecklberger",
"Johannes",
""
],
[
"Lin",
"Wei",
""
],
[
"Hermosilla",
"Pedro",
""
],
[
"Doveh",
"Sivan",
""
],
[
"Possegger",
"Horst",
""
],
[
"Mirza",
"M. Jehanzeb",
""
]
] |
2504.14232 | Antoun Yaacoub | Antoun Yaacoub, J\'er\^ome Da-Rugna, Zainab Assaghir | Assessing AI-Generated Questions' Alignment with Cognitive Frameworks in
Educational Assessment | This paper was presented in the 17th Int. Conf. on Computer Science
and Information Technology (ICCSIT 2024), Dubai, United Arab Emirates, 2024,
Oct. 23-25. IT's now in production to be published in the International
Journal of Computer Theory and Engineering | null | null | null | cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | This study evaluates the integration of Bloom's Taxonomy into OneClickQuiz,
an Artificial Intelligence (AI) driven plugin for automating Multiple-Choice
Question (MCQ) generation in Moodle. Bloom's Taxonomy provides a structured
framework for categorizing educational objectives into hierarchical cognitive
levels. Our research investigates whether incorporating this taxonomy can
improve the alignment of AI-generated questions with specific cognitive
objectives. We developed a dataset of 3691 questions categorized according to
Bloom's levels and employed various classification models-Multinomial Logistic
Regression, Naive Bayes, Linear Support Vector Classification (SVC), and a
Transformer-based model (DistilBERT)-to evaluate their effectiveness in
categorizing questions. Our results indicate that higher Bloom's levels
generally correlate with increased question length, Flesch-Kincaid Grade Level
(FKGL), and Lexical Density (LD), reflecting the increased complexity of higher
cognitive demands. Multinomial Logistic Regression showed varying accuracy
across Bloom's levels, performing best for "Knowledge" and less accurately for
higher-order levels. Merging higher-level categories improved accuracy for
complex cognitive tasks. Naive Bayes and Linear SVC also demonstrated effective
classification for lower levels but struggled with higher-order tasks.
DistilBERT achieved the highest performance, significantly improving
classification of both lower and higher-order cognitive levels, achieving an
overall validation accuracy of 91%. This study highlights the potential of
integrating Bloom's Taxonomy into AI-driven assessment tools and underscores
the advantages of advanced models like DistilBERT for enhancing educational
content generation.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:03:39 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Yaacoub",
"Antoun",
""
],
[
"Da-Rugna",
"Jérôme",
""
],
[
"Assaghir",
"Zainab",
""
]
] |
2504.14233 | Yong-En Tian | Yong-En Tian, Yu-Chien Tang, Kuang-Da Wang, An-Zi Yen, Wen-Chih Peng | Template-Based Financial Report Generation in Agentic and Decomposed
Information Retrieval | 5 pages; 3 figures. Accepted by SIGIR 2025 short paper track. Code
available at
https://github.com/bryant-nn/Template-Based-Financial-Report-Generation | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Tailoring structured financial reports from companies' earnings releases is
crucial for understanding financial performance and has been widely adopted in
real-world analytics. However, existing summarization methods often generate
broad, high-level summaries, which may lack the precision and detail required
for financial reports that typically focus on specific, structured sections.
While Large Language Models (LLMs) hold promise, generating reports adhering to
predefined multi-section templates remains challenging. This paper investigates
two LLM-based approaches popular in industry for generating templated financial
reports: an agentic information retrieval (IR) framework and a decomposed IR
approach, namely AgenticIR and DecomposedIR. The AgenticIR utilizes
collaborative agents prompted with the full template. In contrast, the
DecomposedIR approach applies a prompt chaining workflow to break down the
template and reframe each section as a query answered by the LLM using the
earnings release. To quantitatively assess the generated reports, we evaluated
both methods in two scenarios: one using a financial dataset without direct
human references, and another with a weather-domain dataset featuring
expert-written reports. Experimental results show that while AgenticIR may
excel in orchestrating tasks and generating concise reports through agent
collaboration, DecomposedIR statistically significantly outperforms AgenticIR
approach in providing broader and more detailed coverage in both scenarios,
offering reflection on the utilization of the agentic framework in real-world
applications.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:06:13 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Tian",
"Yong-En",
""
],
[
"Tang",
"Yu-Chien",
""
],
[
"Wang",
"Kuang-Da",
""
],
[
"Yen",
"An-Zi",
""
],
[
"Peng",
"Wen-Chih",
""
]
] |
2504.14234 | Ming Li | Ming Li, Yaru Dong, Ruiqi Zhang, Xuena Zhu, Shuqian Shen, Lei Li, and
Shao-Ming Fei | A note on the lower bounds of genuine multipartite entanglement
concurrence | 15 pages, 2 figures | Quantum Information Processing (2024) 23:397 | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum entanglement plays a pivotal role in quantum information processing.
Quantifying quantum entanglement is a challenging and essential research area
within the field. This manuscript explores the relationships between bipartite
entanglement concurrence, multipartite entanglement concurrence, and genuine
multipartite entanglement (GME) concurrence. We derive lower bounds on GME
concurrence from these relationships, demonstrating their superiority over
existing results through rigorous proofs and numerical examples. Additionally,
we investigate the connections between GME concurrence and other entanglement
measures, such as tangle and global negativity, in multipartite quantum
systems.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:09:43 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Li",
"Ming",
""
],
[
"Dong",
"Yaru",
""
],
[
"Zhang",
"Ruiqi",
""
],
[
"Zhu",
"Xuena",
""
],
[
"Shen",
"Shuqian",
""
],
[
"Li",
"Lei",
""
],
[
"Fei",
"Shao-Ming",
""
]
] |
2504.14235 | Irdin Pekaric PhD. | Saskia Schr\"oer, No\'e Canevascini, Irdin Pekaric, Philine Widmer and
Pavel Laskov | The Dark Side of the Web: Towards Understanding Various Data Sources in
Cyber Threat Intelligence | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Cyber threats have become increasingly prevalent and sophisticated. Prior
work has extracted actionable cyber threat intelligence (CTI), such as
indicators of compromise, tactics, techniques, and procedures (TTPs), or threat
feeds from various sources: open source data (e.g., social networks), internal
intelligence (e.g., log data), and ``first-hand'' communications from
cybercriminals (e.g., underground forums, chats, darknet websites). However,
"first-hand" data sources remain underutilized because it is difficult to
access or scrape their data.
In this work, we analyze (i) 6.6 million posts, (ii) 3.4 million messages,
and (iii) 120,000 darknet websites. We combine NLP tools to address several
challenges in analyzing such data. First, even on dedicated platforms, only
some content is CTI-relevant, requiring effective filtering. Second,
"first-hand" data can be CTI-relevant from a technical or strategic viewpoint.
We demonstrate how to organize content along this distinction. Third, we
describe the topics discussed and how "first-hand" data sources differ from
each other. According to our filtering, 20% of our sample is CTI-relevant. Most
of the CTI-relevant data focuses on strategic rather than technical
discussions. Credit card-related crime is the most prevalent topic on darknet
websites. On underground forums and chat channels, account and subscription
selling is discussed most. Topic diversity is higher on underground forums and
chat channels than on darknet websites. Our analyses suggest that different
platforms may be used for activities with varying complexity and risks for
criminals.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:14:10 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Schröer",
"Saskia",
""
],
[
"Canevascini",
"Noé",
""
],
[
"Pekaric",
"Irdin",
""
],
[
"Widmer",
"Philine",
""
],
[
"Laskov",
"Pavel",
""
]
] |
2504.14236 | Anna Galler | Anna Galler, Ofer Neufeld | Bulk photogalvanic current control and gap spectroscopy in 2D hexagonal
materials | null | null | null | null | cond-mat.mtrl-sci physics.optics | http://creativecommons.org/licenses/by/4.0/ | Two-dimensional (2D) hexagonal materials have been intensively explored for
multiple optoelectronic applications such as spin current generation,
all-optical valleytronics, and topological electronics. In the realm of
strong-field and ultrafast light-driven phenomena, it was shown that tailored
laser driving such as polychromatic or few-cycle pulses can drive robust bulk
photogalvanic (BPG) currents originating from the K/K' valleys. We here explore
the BPG effect in 2D systems in the strong-field regime and show that
monochromatic elliptical pulses also generically generate such photocurrents.
The resultant photocurrents exhibit both parallel and transverse (Hall-like)
components, both highly sensitive to the laser parameters, providing
photocurrent control knobs. Interestingly, we show that the photocurrent
amplitude has a distinct behavior vs. the driving ellipticity that can be
indicative of material properties such as the gap size at K/K', which should
prove useful for novel forms of BPG-based spectroscopies. We demonstrate these
effects also in benchmark ab-initio simulations in monolayer hexagonal
boron-nitride. Our work establishes new paths for controlling photocurrent
responses in 2D systems that can also be used for multi-dimensional
spectroscopy of ultrafast material properties through photocurrent
measurements.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:14:57 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Galler",
"Anna",
""
],
[
"Neufeld",
"Ofer",
""
]
] |
2504.14237 | Dekang Zhang | Dekang Zhang, Dan Niu, Zhou Jin, Yichao Dong, Jingweijia Tan, Changyin
Sun | A Novel Frequency-Spatial Domain Aware Network for Fast Thermal
Prediction in 2.5D ICs | 7 pages, 5 figures, 22nd Design, Automation and Test in Europe
Conference (DATE '25) | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In the post-Moore era, 2.5D chiplet-based ICs present significant challenges
in thermal management due to increased power density and thermal hotspots.
Neural network-based thermal prediction models can perform real-time
predictions for many unseen new designs. However, existing CNN-based and
GCN-based methods cannot effectively capture the global thermal features,
especially for high-frequency components, hindering prediction accuracy
enhancement. In this paper, we propose a novel frequency-spatial dual domain
aware prediction network (FSA-Heat) for fast and high-accuracy thermal
prediction in 2.5D ICs. It integrates high-to-low frequency and spatial domain
encoder (FSTE) module with frequency domain cross-scale interaction module
(FCIFormer) to achieve high-to-low frequency and global-to-local thermal
dissipation feature extraction. Additionally, a frequency-spatial hybrid loss
(FSL) is designed to effectively attenuate high-frequency thermal gradient
noise and spatial misalignments. The experimental results show that the
performance enhancements offered by our proposed method are substantial,
outperforming the newly-proposed 2.5D method, GCN+PNA, by considerable margins
(over 99% RMSE reduction, 4.23X inference time speedup). Moreover, extensive
experiments demonstrate that FSA-Heat also exhibits robust generalization
capabilities.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:16:30 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Dekang",
""
],
[
"Niu",
"Dan",
""
],
[
"Jin",
"Zhou",
""
],
[
"Dong",
"Yichao",
""
],
[
"Tan",
"Jingweijia",
""
],
[
"Sun",
"Changyin",
""
]
] |
2504.14238 | Lu Pan | Lu Pan, Yu-Hsuan Huang, Hongxia Xie, Cheng Zhang, Hongwei Zhao,
Hong-Han Shuai, Wen-Huang Cheng | Single Document Image Highlight Removal via A Large-Scale Real-World
Dataset and A Location-Aware Network | main paper with 8 pages, conference | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reflective documents often suffer from specular highlights under ambient
lighting, severely hindering text readability and degrading overall visual
quality. Although recent deep learning methods show promise in highlight
removal, they remain suboptimal for document images, primarily due to the lack
of dedicated datasets and tailored architectural designs. To tackle these
challenges, we present DocHR14K, a large-scale real-world dataset comprising
14,902 high-resolution image pairs across six document categories and various
lighting conditions. To the best of our knowledge, this is the first
high-resolution dataset for document highlight removal that captures a wide
range of real-world lighting conditions. Additionally, motivated by the
observation that the residual map between highlighted and clean images
naturally reveals the spatial structure of highlight regions, we propose a
simple yet effective Highlight Location Prior (HLP) to estimate highlight masks
without human annotations. Building on this prior, we present the
Location-Aware Laplacian Pyramid Highlight Removal Network (L2HRNet), which
effectively removes highlights by leveraging estimated priors and incorporates
diffusion module to restore details. Extensive experiments demonstrate that
DocHR14K improves highlight removal under diverse lighting conditions. Our
L2HRNet achieves state-of-the-art performance across three benchmark datasets,
including a 5.01\% increase in PSNR and a 13.17\% reduction in RMSE on
DocHR14K.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:20:11 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Pan",
"Lu",
""
],
[
"Huang",
"Yu-Hsuan",
""
],
[
"Xie",
"Hongxia",
""
],
[
"Zhang",
"Cheng",
""
],
[
"Zhao",
"Hongwei",
""
],
[
"Shuai",
"Hong-Han",
""
],
[
"Cheng",
"Wen-Huang",
""
]
] |
2504.14239 | Yuhang Liu | Yuhang Liu, Pengxiang Li, Congkai Xie, Xavier Hu, Xiaotian Han,
Shengyu Zhang, Hongxia Yang, Fei Wu | InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to
Deliberative Reasoners | 10 pages, 3 figures, work in progress | null | null | null | cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Multimodal Large Language Models (MLLMs) have powered Graphical User
Interface (GUI) Agents, showing promise in automating tasks on computing
devices. Recent works have begun exploring reasoning in GUI tasks with
encouraging results. However, many current approaches rely on manually designed
reasoning templates, which may result in reasoning that is not sufficiently
robust and adaptive for complex GUI environments. Meanwhile, some existing
agents continue to operate as Reactive Actors, relying primarily on implicit
reasoning that may lack sufficient depth for GUI tasks demanding planning and
error recovery. We argue that advancing these agents requires a shift from
reactive acting towards acting based on deliberate reasoning. To facilitate
this transformation, we introduce InfiGUI-R1, an MLLM-based GUI agent developed
through our Actor2Reasoner framework, a reasoning-centric, two-stage training
approach designed to progressively evolve agents from Reactive Actors to
Deliberative Reasoners. The first stage, Reasoning Injection, focuses on
establishing a basic reasoner. We employ Spatial Reasoning Distillation to
transfer cross-modal spatial reasoning capabilities from teacher models to
MLLMs through trajectories with explicit reasoning steps, enabling models to
integrate GUI visual-spatial information with logical reasoning before action
generation. The second stage, Deliberation Enhancement, refines the basic
reasoner into a deliberative one using Reinforcement Learning. This stage
introduces two approaches: Sub-goal Guidance, which rewards models for
generating accurate intermediate sub-goals, and Error Recovery Scenario
Construction, which creates failure-and-recovery training scenarios from
identified prone-to-error steps. Experimental results show InfiGUI-R1 achieves
strong performance in GUI grounding and trajectory tasks. Resources at
https://github.com/Reallm-Labs/InfiGUI-R1.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:25:55 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Liu",
"Yuhang",
""
],
[
"Li",
"Pengxiang",
""
],
[
"Xie",
"Congkai",
""
],
[
"Hu",
"Xavier",
""
],
[
"Han",
"Xiaotian",
""
],
[
"Zhang",
"Shengyu",
""
],
[
"Yang",
"Hongxia",
""
],
[
"Wu",
"Fei",
""
]
] |
2504.14240 | Liang Xie | Xie Liang and Gao Wei and Zhenghui Ming and Li Ge | ROI-Guided Point Cloud Geometry Compression Towards Human and Machine
Vision | 10 pages, 5 figures | ACM International Conference on Multimedia 2024 | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Point cloud data is pivotal in applications like autonomous driving, virtual
reality, and robotics. However, its substantial volume poses significant
challenges in storage and transmission. In order to obtain a high compression
ratio, crucial semantic details usually confront severe damage, leading to
difficulties in guaranteeing the accuracy of downstream tasks. To tackle this
problem, we are the first to introduce a novel Region of Interest (ROI)-guided
Point Cloud Geometry Compression (RPCGC) method for human and machine vision.
Our framework employs a dual-branch parallel structure, where the base layer
encodes and decodes a simplified version of the point cloud, and the
enhancement layer refines this by focusing on geometry details. Furthermore,
the residual information of the enhancement layer undergoes refinement through
an ROI prediction network. This network generates mask information, which is
then incorporated into the residuals, serving as a strong supervision signal.
Additionally, we intricately apply these mask details in the Rate-Distortion
(RD) optimization process, with each point weighted in the distortion
calculation. Our loss function includes RD loss and detection loss to better
guide point cloud encoding for the machine. Experiment results demonstrate that
RPCGC achieves exceptional compression performance and better detection
accuracy (10% gain) than some learning-based compression methods at high
bitrates in ScanNet and SUN RGB-D datasets.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:31:37 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Liang",
"Xie",
""
],
[
"Wei",
"Gao",
""
],
[
"Ming",
"Zhenghui",
""
],
[
"Ge",
"Li",
""
]
] |
2504.14241 | Chengming Wang | Chengming Wang, Dongyao Jia, Wei Wang, Dong Ngoduy, Bei Peng, Jianping
Wang | A Knowledge-Informed Deep Learning Paradigm for Generalizable and
Stability-Optimized Car-Following Models | null | null | null | null | cs.AI cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Car-following models (CFMs) are fundamental to traffic flow analysis and
autonomous driving. Although calibrated physics-based and trained data-driven
CFMs can replicate human driving behavior, their reliance on specific datasets
limits generalization across diverse scenarios and reduces reliability in
real-world deployment. Moreover, these models typically focus on behavioral
fidelity and do not support the explicit optimization of local and string
stability, which are increasingly important for the safe and efficient
operation of autonomous vehicles (AVs). To address these limitations, we
propose a Knowledge-Informed Deep Learning (KIDL) paradigm that distills the
generalization capabilities of pre-trained Large Language Models (LLMs) into a
lightweight and stability-aware neural architecture. LLMs are used to extract
fundamental car-following knowledge beyond dataset-specific patterns, and this
knowledge is transferred to a reliable, tractable, and computationally
efficient model through knowledge distillation. KIDL also incorporates
stability constraints directly into its training objective, ensuring that the
resulting model not only emulates human-like behavior but also satisfies the
local and string stability requirements essential for real-world AV deployment.
We evaluate KIDL on the real-world NGSIM and HighD datasets, comparing its
performance with representative physics-based, data-driven, and hybrid CFMs.
Both empirical and theoretical results consistently demonstrate KIDL's superior
behavioral generalization and traffic flow stability, offering a robust and
scalable solution for next-generation traffic systems.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:33:02 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Wang",
"Chengming",
""
],
[
"Jia",
"Dongyao",
""
],
[
"Wang",
"Wei",
""
],
[
"Ngoduy",
"Dong",
""
],
[
"Peng",
"Bei",
""
],
[
"Wang",
"Jianping",
""
]
] |
2504.14242 | Sepideh Ghaziasgar | Sepideh Ghaziasgar, Mahdi Abdollahi, Atefeh Javadi, Jacco Th. van
Loon, Iain McDonald, Joana Oliveira, Habib G. Khosroshahi | Machine Learning Classification of Young Stellar Objects and Evolved
Stars in the Magellanic Clouds Using the Probabilistic Random Forest
Classifier | 6 pages, 2 figures, 2 tables, Published in the Communications of BAO
(ComBAO) | null | 10.52526/25792776-24.71.2-377 | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | The Magellanic Clouds (MCs) are excellent locations to study stellar dust
emission and its contribution to galaxy evolution. Through spectral and
photometric classification, MCs can serve as a unique environment for studying
stellar evolution and galaxies enriched by dusty stellar point sources. We
applied machine learning classifiers to spectroscopically labeled data from the
Surveying the Agents of Galaxy Evolution (SAGE) project, which involved 12
multiwavelength filters and 618 stellar objects at the MCs. We classified stars
into five categories: young stellar objects (YSOs), carbon-rich asymptotic
giant branch (CAGB) stars, oxygen-rich AGB (OAGB) stars, red supergiants (RSG),
and post-AGB (PAGB) stars. Following this, we augmented the distribution of
imbalanced classes using the Synthetic Minority Oversampling Technique (SMOTE).
Therefore, the Probabilistic Random Forest (PRF) classifier achieved the
highest overall accuracy, reaching ${89\%}$ based on the recall metric, in
categorizing dusty stellar sources before and after data augmentation. In this
study, SMOTE did not impact the classification accuracy for the CAGB, PAGB, and
RSG categories but led to changes in the performance of the OAGB and YSO
classes.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:34:38 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ghaziasgar",
"Sepideh",
""
],
[
"Abdollahi",
"Mahdi",
""
],
[
"Javadi",
"Atefeh",
""
],
[
"van Loon",
"Jacco Th.",
""
],
[
"McDonald",
"Iain",
""
],
[
"Oliveira",
"Joana",
""
],
[
"Khosroshahi",
"Habib G.",
""
]
] |
2504.14243 | Yimeng Bai | Yimeng Bai, Shunyu Zhang, Yang Zhang, Hu Liu, Wentian Bao, Enyun Yu,
Fuli Feng, Wenwu Ou | Unconstrained Monotonic Calibration of Predictions in Deep Ranking
Systems | Accepted by SIGIR'25 | null | 10.1145/3726302.3730105 | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ranking models primarily focus on modeling the relative order of predictions
while often neglecting the significance of the accuracy of their absolute
values. However, accurate absolute values are essential for certain downstream
tasks, necessitating the calibration of the original predictions. To address
this, existing calibration approaches typically employ predefined
transformation functions with order-preserving properties to adjust the
original predictions. Unfortunately, these functions often adhere to fixed
forms, such as piece-wise linear functions, which exhibit limited
expressiveness and flexibility, thereby constraining their effectiveness in
complex calibration scenarios. To mitigate this issue, we propose implementing
a calibrator using an Unconstrained Monotonic Neural Network (UMNN), which can
learn arbitrary monotonic functions with great modeling power. This approach
significantly relaxes the constraints on the calibrator, improving its
flexibility and expressiveness while avoiding excessively distorting the
original predictions by requiring monotonicity. Furthermore, to optimize this
highly flexible network for calibration, we introduce a novel additional loss
function termed Smooth Calibration Loss (SCLoss), which aims to fulfill a
necessary condition for achieving the ideal calibration state. Extensive
offline experiments confirm the effectiveness of our method in achieving
superior calibration performance. Moreover, deployment in Kuaishou's
large-scale online video ranking system demonstrates that the method's
calibration improvements translate into enhanced business metrics. The source
code is available at https://github.com/baiyimeng/UMC.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:35:11 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Bai",
"Yimeng",
""
],
[
"Zhang",
"Shunyu",
""
],
[
"Zhang",
"Yang",
""
],
[
"Liu",
"Hu",
""
],
[
"Bao",
"Wentian",
""
],
[
"Yu",
"Enyun",
""
],
[
"Feng",
"Fuli",
""
],
[
"Ou",
"Wenwu",
""
]
] |
2504.14244 | Boris Kerner | Boris S. Kerner and Sergey L. Klenov | Microscopic features of the effect of vehicle overacceleration on
traffic flow | 28 pages, 32 figures | null | null | null | physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | Through the development of a microscopic deterministic model in the framework
of three-phase traffic theory, microscopic features of vehicle
overacceleration, which determines the occurrence of the metastability of free
traffic flow at a bottleneck, have been revealed: (i) The greater the impact of
vehicle overacceleration on free traffic flow at a bottleneck, the higher the
maximum flow rate at which free flow can persist at the bottleneck, i.e., the
better traffic breakdown can be avoided. (ii) There can be at least two
mechanisms of overacceleration in road lane caused by safety acceleration at
the bottleneck. (iii) Through a microscopic analysis of spatiotemporal
competition between speed adaptation and vehicle acceleration behaviors,
traffic conditions have been found at which safety acceleration in road lane
or/and vehicle acceleration due to lane-changing on multi-lane road become
overacceleration. (iv) There is spatiotemporal cooperation of different
overacceleration mechanisms. (v) The stronger the overacceleration cooperation,
the stronger the maintenance of free flow at the bottleneck due to
overacceleration. (vi) On two-lane road, both speed adaptation and
overacceleration in road lane can effect qualitatively on the overacceleration
mechanism caused by lane-changing. These microscopic features of the effect of
vehicle overacceleration on traffic flow are related to traffic flow consisting
of human-driving or/and automated-driving vehicles.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:37:17 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Kerner",
"Boris S.",
""
],
[
"Klenov",
"Sergey L.",
""
]
] |
2504.14245 | Yikun Ji | Yikun Ji, Yan Hong, Jiahui Zhan, Haoxing Chen, jun lan, Huijia Zhu,
Weiqiang Wang, Liqing Zhang, Jianfu Zhang | Towards Explainable Fake Image Detection with Multi-Modal Large Language
Models | null | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Progress in image generation raises significant public security concerns. We
argue that fake image detection should not operate as a "black box". Instead,
an ideal approach must ensure both strong generalization and transparency.
Recent progress in Multi-modal Large Language Models (MLLMs) offers new
opportunities for reasoning-based AI-generated image detection. In this work,
we evaluate the capabilities of MLLMs in comparison to traditional detection
methods and human evaluators, highlighting their strengths and limitations.
Furthermore, we design six distinct prompts and propose a framework that
integrates these prompts to develop a more robust, explainable, and
reasoning-driven detection system. The code is available at
https://github.com/Gennadiyev/mllm-defake.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:42:25 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ji",
"Yikun",
""
],
[
"Hong",
"Yan",
""
],
[
"Zhan",
"Jiahui",
""
],
[
"Chen",
"Haoxing",
""
],
[
"lan",
"jun",
""
],
[
"Zhu",
"Huijia",
""
],
[
"Wang",
"Weiqiang",
""
],
[
"Zhang",
"Liqing",
""
],
[
"Zhang",
"Jianfu",
""
]
] |
2504.14246 | Jinbang Yang | Zhenmou Liu and Jinbang Yang and Kang Zuo | Logarithmic Crystalline Representations | 21 pages | null | null | null | math.AG math.RT | http://creativecommons.org/licenses/by/4.0/ | In 1989, Faltings proved the comparison theorem between \'etale cohomology
and crystalline cohomology by studying Fontaine-Faltings modules and
crystalline representations. In his paper, he mentioned these modules and
representations can be extended to the logarithmic context, but without detail.
This note aims to explicitly present the construction of logarithmic
Fontaine-Faltings modules and logarithmic crystalline representations.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:42:44 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Liu",
"Zhenmou",
""
],
[
"Yang",
"Jinbang",
""
],
[
"Zuo",
"Kang",
""
]
] |
2504.14247 | Malcolm Perry | Mirjam Cvetic and Malcolm J. Perry | Quantization of Black Hole Entropy for Black Holes in Subtracted
Geometry | 6 pages | null | null | null | hep-th gr-qc | http://creativecommons.org/licenses/by/4.0/ | We carefully examine the exact analytic spectrum of quasinormal modes of
general black holes in the so-called subtracted geometry of maximally
supersymmetric supergravity. These black holes have the same area and surface
gravity at both the outer and inner horizons as the original asymptotically
flat black holes. We proceed to explore the relationship with conformal field
theories that describe horizon physics of these black holes. As a consequence,
we show that the horizon area of these black holes is quantized in units of
8{\pi}l_{Planck}^2.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:45:04 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Cvetic",
"Mirjam",
""
],
[
"Perry",
"Malcolm J.",
""
]
] |
2504.14248 | Shijiao Li | Li Shijiao, Ma Zhipeng, He Huajun, Chen Haiyue | Rethinking Traffic Flow Forecasting: From Transition to Generatation | null | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Traffic flow prediction plays an important role in Intelligent Transportation
Systems in traffic management and urban planning. There have been extensive
successful works in this area. However, these approaches focus only on
modelling the flow transition and ignore the flow generation process, which
manifests itself in two ways: (i) The models are based on Markovian
assumptions, ignoring the multi-periodicity of the flow generation in nodes.
(ii) The same structure is designed to encode both the transition and
generation processes, ignoring the differences between them. To address these
problems, we propose an Effective Multi-Branch Similarity Transformer for
Traffic Flow Prediction, namely EMBSFormer. Through data analysis, we find that
the factors affecting traffic flow include node-level traffic generation and
graph-level traffic transition, which describe the multi-periodicity and
interaction pattern of nodes, respectively. Specifically, to capture traffic
generation patterns, we propose a similarity analysis module that supports
multi-branch encoding to dynamically expand significant cycles. For traffic
transition, we employ a temporal and spatial self-attention mechanism to
maintain global node interactions, and use GNN and time conv to model local
node interactions, respectively. Model performance is evaluated on three
real-world datasets on both long-term and short-term prediction tasks.
Experimental results show that EMBSFormer outperforms baselines on both tasks.
Moreover, compared to models based on flow transition modelling (e.g. GMAN,
513k), the variant of EMBSFormer(93K) only uses 18\% of the parameters,
achieving the same performance.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:52:39 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Shijiao",
"Li",
""
],
[
"Zhipeng",
"Ma",
""
],
[
"Huajun",
"He",
""
],
[
"Haiyue",
"Chen",
""
]
] |
2504.14249 | Bin Ren | Bin Ren and Eduard Zamfir and Zongwei Wu and Yawei Li and Yidi Li and
Danda Pani Paudel and Radu Timofte and Ming-Hsuan Yang and Luc Van Gool and
Nicu Sebe | Any Image Restoration via Efficient Spatial-Frequency Degradation
Adaptation | Efficient All in One Image Restoration | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Restoring any degraded image efficiently via just one model has become
increasingly significant and impactful, especially with the proliferation of
mobile devices. Traditional solutions typically involve training dedicated
models per degradation, resulting in inefficiency and redundancy. More recent
approaches either introduce additional modules to learn visual prompts,
significantly increasing model size, or incorporate cross-modal transfer from
large language models trained on vast datasets, adding complexity to the system
architecture. In contrast, our approach, termed AnyIR, takes a unified path
that leverages inherent similarity across various degradations to enable both
efficient and comprehensive restoration through a joint embedding mechanism,
without scaling up the model or relying on large language models.Specifically,
we examine the sub-latent space of each input, identifying key components and
reweighting them first in a gated manner. To fuse the intrinsic degradation
awareness and the contextualized attention, a spatial-frequency parallel fusion
strategy is proposed for enhancing spatial-aware local-global interactions and
enriching the restoration details from the frequency perspective. Extensive
benchmarking in the all-in-one restoration setting confirms AnyIR's SOTA
performance, reducing model complexity by around 82\% in parameters and 85\% in
FLOPs. Our code will be available at our Project page
(https://amazingren.github.io/AnyIR/)
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:54:46 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ren",
"Bin",
""
],
[
"Zamfir",
"Eduard",
""
],
[
"Wu",
"Zongwei",
""
],
[
"Li",
"Yawei",
""
],
[
"Li",
"Yidi",
""
],
[
"Paudel",
"Danda Pani",
""
],
[
"Timofte",
"Radu",
""
],
[
"Yang",
"Ming-Hsuan",
""
],
[
"Van Gool",
"Luc",
""
],
[
"Sebe",
"Nicu",
""
]
] |
2504.14250 | Yunhui Liu | Yunhui Liu, Jiashun Cheng, Jia Li, Fugee Tsung, Hongzhi Yin, Tieke He | A Pre-Training and Adaptive Fine-Tuning Framework for Graph Anomaly
Detection | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph anomaly detection (GAD) has garnered increasing attention in recent
years, yet it remains challenging due to the scarcity of abnormal nodes and the
high cost of label annotations. Graph pre-training, the two-stage learning
paradigm, has emerged as an effective approach for label-efficient learning,
largely benefiting from expressive neighborhood aggregation under the
assumption of strong homophily. However, in GAD, anomalies typically exhibit
high local heterophily, while normal nodes retain strong homophily, resulting
in a complex homophily-heterophily mixture. To understand the impact of this
mixed pattern on graph pre-training, we analyze it through the lens of spectral
filtering and reveal that relying solely on a global low-pass filter is
insufficient for GAD. We further provide a theoretical justification for the
necessity of selectively applying appropriate filters to individual nodes.
Building upon this insight, we propose PAF, a Pre-Training and Adaptive
Fine-tuning framework specifically designed for GAD. In particular, we
introduce joint training with low- and high-pass filters in the pre-training
phase to capture the full spectrum of frequency information in node features.
During fine-tuning, we devise a gated fusion network that adaptively combines
node representations generated by both filters. Extensive experiments across
ten benchmark datasets consistently demonstrate the effectiveness of PAF.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:57:35 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Liu",
"Yunhui",
""
],
[
"Cheng",
"Jiashun",
""
],
[
"Li",
"Jia",
""
],
[
"Tsung",
"Fugee",
""
],
[
"Yin",
"Hongzhi",
""
],
[
"He",
"Tieke",
""
]
] |
2504.14251 | Mirko Giacchini | Flavio Chierichetti, Mirko Giacchini, Alessandro Panconesi, Andrea
Vattani | A New Impossibility Result for Online Bipartite Matching Problems | To appear in Proceedings of ICALP 2025 | null | null | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Online Bipartite Matching with random user arrival is a fundamental problem
in the online advertisement ecosystem. Over the last 30 years, many algorithms
and impossibility results have been developed for this problem. In particular,
the latest impossibility result was established by Manshadi, Oveis Gharan and
Saberi in 2011. Since then, several algorithms have been published in an effort
to narrow the gap between the upper and the lower bounds on the competitive
ratio.
In this paper we show that no algorithm can achieve a competitive ratio
better than $1- \frac e{e^e} = 0.82062\ldots$, improving upon the $0.823$ upper
bound presented in (Manshadi, Oveis Gharan and Saberi, SODA 2011). Our
construction is simple to state, accompanied by a fully analytic proof, and
yields a competitive ratio bound intriguingly similar to $1 - \frac1e$, the
optimal competitive ratio for the fully adversarial Online Bipartite Matching
problem.
Although the tightness of our upper bound remains an open question, we show
that our construction is extremal in a natural class of instances.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 09:59:40 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Chierichetti",
"Flavio",
""
],
[
"Giacchini",
"Mirko",
""
],
[
"Panconesi",
"Alessandro",
""
],
[
"Vattani",
"Andrea",
""
]
] |
2504.14252 | Jan Heuer | Jan Heuer | Ordered Completion for Non-Locally Tight mini-gringo Programs | Master's Thesis submitted at the University of Potsdam | null | null | null | cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Completion is a well-known transformation that captures the stable model
semantics of logic programs by turning a program into a set of first-order
definitions. Stable models are models of the completion, but not all models of
the completion are stable models. For tight programs (programs without positive
recursion) the two semantics coincide. Recently this correspondence was
extended to locally tight programs, which avoid non-terminating recursion.
However, unlike tightness, local tightness cannot be checked with simple
syntactic methods. Completion is crucial for verifying answer set programs,
especially for external equivalence: a form of equivalence based on selected
output predicates under certain inputs. Standard equivalence and adherence to a
first-order specification are special cases of external equivalence. The anthem
verification tool has two limitations for checking external equivalence: (1)
there is no way to check local tightness automatically, and (2) it is not
possible to verify programs that are not locally tight. Therefore, alternatives
to completion are of interest. This thesis investigates ordered completion,
introduced in [Asuncion et al., 2012], which captures stable models of
arbitrary logic programs, but only for finite models. This work extends ordered
completion to the mini-gringo language (a subset of the language used by the
clingo solver). Additionally, it introduces a modification of ordered
completion to handle infinite stable models. This extended ordered completion
is implemented in anthem as a translation, and initial experiments demonstrate
its use for verifying simple logic programs.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:00:50 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Heuer",
"Jan",
""
]
] |
2504.14253 | Yifan Wang | Yifan Wang, Jie Gui, Xinli Shi, Linqing Gui, Yuan Yan Tang, James
Tin-Yau Kwok | ColorVein: Colorful Cancelable Vein Biometrics | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Vein recognition technologies have become one of the primary solutions for
high-security identification systems. However, the issue of biometric
information leakage can still pose a serious threat to user privacy and
anonymity. Currently, there is no cancelable biometric template generation
scheme specifically designed for vein biometrics. Therefore, this paper
proposes an innovative cancelable vein biometric generation scheme: ColorVein.
Unlike previous cancelable template generation schemes, ColorVein does not
destroy the original biometric features and introduces additional color
information to grayscale vein images. This method significantly enhances the
information density of vein images by transforming static grayscale information
into dynamically controllable color representations through interactive
colorization. ColorVein allows users/administrators to define a controllable
pseudo-random color space for grayscale vein images by editing the position,
number, and color of hint points, thereby generating protected cancelable
templates. Additionally, we propose a new secure center loss to optimize the
training process of the protected feature extraction model, effectively
increasing the feature distance between enrolled users and any potential
impostors. Finally, we evaluate ColorVein's performance on all types of vein
biometrics, including recognition performance, unlinkability, irreversibility,
and revocability, and conduct security and privacy analyses. ColorVein achieves
competitive performance compared with state-of-the-art methods.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:03:49 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Wang",
"Yifan",
""
],
[
"Gui",
"Jie",
""
],
[
"Shi",
"Xinli",
""
],
[
"Gui",
"Linqing",
""
],
[
"Tang",
"Yuan Yan",
""
],
[
"Kwok",
"James Tin-Yau",
""
]
] |
2504.14254 | Jie Wang | Jie Wang, Nana Yu, Zihao Zhang, Yahong Han | Visual Consensus Prompting for Co-Salient Object Detection | CVPR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing co-salient object detection (CoSOD) methods generally employ a
three-stage architecture (i.e., encoding, consensus extraction & dispersion,
and prediction) along with a typical full fine-tuning paradigm. Although they
yield certain benefits, they exhibit two notable limitations: 1) This
architecture relies on encoded features to facilitate consensus extraction, but
the meticulously extracted consensus does not provide timely guidance to the
encoding stage. 2) This paradigm involves globally updating all parameters of
the model, which is parameter-inefficient and hinders the effective
representation of knowledge within the foundation model for this task.
Therefore, in this paper, we propose an interaction-effective and
parameter-efficient concise architecture for the CoSOD task, addressing two key
limitations. It introduces, for the first time, a parameter-efficient prompt
tuning paradigm and seamlessly embeds consensus into the prompts to formulate
task-specific Visual Consensus Prompts (VCP). Our VCP aims to induce the frozen
foundation model to perform better on CoSOD tasks by formulating task-specific
visual consensus prompts with minimized tunable parameters. Concretely, the
primary insight of the purposeful Consensus Prompt Generator (CPG) is to
enforce limited tunable parameters to focus on co-salient representations and
generate consensus prompts. The formulated Consensus Prompt Disperser (CPD)
leverages consensus prompts to form task-specific visual consensus prompts,
thereby arousing the powerful potential of pre-trained models in addressing
CoSOD tasks. Extensive experiments demonstrate that our concise VCP outperforms
13 cutting-edge full fine-tuning models, achieving the new state of the art
(with 6.8% improvement in F_m metrics on the most challenging CoCA dataset).
Source code has been available at https://github.com/WJ-CV/VCP.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:12:39 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Wang",
"Jie",
""
],
[
"Yu",
"Nana",
""
],
[
"Zhang",
"Zihao",
""
],
[
"Han",
"Yahong",
""
]
] |
2504.14255 | Tomoki Matsuoka | Tomoki Matsuoka, Keiichi Maeda, and Ke-Jung Chen | Inferring Dense Confined Circumstellar Medium around Supernova
Progenitors via Long-term Hydrodynamical Evolution | 10 pages, 5 figures, 1 table. Submitted to The Astrophysical Journal
Letter | null | null | null | astro-ph.HE astro-ph.GA astro-ph.SR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Circumstellar interaction of supernova (SN) ejecta is an essential process in
its evolution and observations of SNe have found the signature of circumstellar
interaction both in the early and late evolutionary phase of SNe. In this
Letter, we show that if the SN forward shock plunges into tenuous stellar wind
from dense circumstellar medium (CSM) in the vicinity of the progenitor (i.e.,
confined CSM), the subsequent time evolutions of the SN-CSM interaction system
deviates from the prediction of self-similar solution. In this case, after all
of the confined CSM is swept up by the SN forward shock (roughly $10$ days
after the explosion), the propagation of the shocked shell will be driven by
the freely expanding ram pressure of the confined CSM component, instead of the
SN ejecta. Meanwhile, the forward shock decelerates faster than the prediction
of thin-shell approximation once the confined CSM component reaches homologous
expansion. This lasts until the reverse shock in the confined CSM component
reaches the head of the SN ejecta, leading to the restoration of the system
into the evolutionary model without confined CSM, where the SN ejecta drives
the expansion of the system. We also show that this peculiar evolution will be
reflected in observational signatures originating from SN-CSM interaction,
taking rapid decline and rebrightening of radio emission as examples. Our
results shed light on the importance of taking into account the effect of
initial SN-CSM interaction even when we focus on observational properties of
SNe a few years after the explosion.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:16:06 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Matsuoka",
"Tomoki",
""
],
[
"Maeda",
"Keiichi",
""
],
[
"Chen",
"Ke-Jung",
""
]
] |
2504.14256 | Florian Galliot | Florian Galliot, Jonas S\'enizergues | Maker-Maker games of rank 4 are PSPACE-complete | null | null | null | null | cs.DM cs.CC math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Maker-Maker convention of positional games is played on a hypergraph
whose edges are interpreted as winning sets. Two players take turns picking a
previously unpicked vertex, aiming at being first to pick all the vertices of
some edge. Optimal play can only lead to a first player win or a draw, and
deciding between the two is known to be PSPACE-complete even for 6-uniform
hypergraphs. We establish PSPACE-completeness for hypergraphs of rank 4. As an
intermediary, we use the recently introduced achievement positional games, a
more general convention in which each player has their own winning sets (blue
and red). We show that deciding whether the blue player has a winning strategy
as the first player is PSPACE-complete even with blue edges of size 2 or 3 and
pairwise disjoint red edges of size 2. The result for hypergraphs of rank 4 in
the Maker-Maker convention follows as a simple corollary.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:20:40 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Galliot",
"Florian",
""
],
[
"Sénizergues",
"Jonas",
""
]
] |
2504.14258 | Nils Morawietz | George B. Mertzios, Hendrik Molter, Nils Morawietz, Paul G. Spirakis | Temporal Graph Realization With Bounded Stretch | null | null | null | null | cs.DS cs.DM | http://creativecommons.org/licenses/by/4.0/ | A periodic temporal graph, in its simplest form, is a graph in which every
edge appears exactly once in the first $\Delta$ time steps, and then it
reappears recurrently every $\Delta$ time steps, where $\Delta$ is a given
period length. This model offers a natural abstraction of transportation
networks where each transportation link connects two destinations periodically.
From a network design perspective, a crucial task is to assign the time-labels
on the edges in a way that optimizes some criterion. In this paper we introduce
a very natural optimality criterion that captures how the temporal distances of
all vertex pairs are `stretched', compared to their physical distances, i.e.
their distances in the underlying static (non-temporal) graph. Given a static
graph $G$, the task is to assign to each edge one time-label between 1 and
$\Delta$ such that, in the resulting periodic temporal graph with
period~$\Delta$, the duration of the fastest temporal path from any vertex $u$
to any other vertex $v$ is at most $\alpha$ times the distance between $u$ and
$v$ in $G$. Here, the value of $\alpha$ measures how much the shortest paths
are allowed to be \emph{stretched} once we assign the periodic time-labels.
Our results span three different directions: First, we provide a series of
approximation and NP-hardness results. Second, we provide approximation and
fixed-parameter algorithms. Among them, we provide a simple polynomial-time
algorithm (the \textit{radius-algorithm}) which always guarantees an
approximation strictly smaller than $\Delta$, and which also computes the
optimum stretch in some cases. Third, we consider a parameterized local search
extension of the problem where we are given the temporal labeling of the graph,
but we are allowed to change the time-labels of at most $k$ edges; for this
problem we prove that it is W[2]-hard but admits an XP algorithm with respect
to $k$.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:35:43 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Mertzios",
"George B.",
""
],
[
"Molter",
"Hendrik",
""
],
[
"Morawietz",
"Nils",
""
],
[
"Spirakis",
"Paul G.",
""
]
] |
2504.14259 | Hadeel Jazzaa PhD | Hadeel Jazzaa, Thomas McCluskey, David Peebles | Experience-based Refinement of Task Planning Knowledge in Autonomous
Robots | null | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | The requirement for autonomous robots to exhibit higher-level cognitive
skills by planning and adapting in an ever-changing environment is indeed a
great challenge for the AI community. Progress has been made in the automated
planning community on refinement and repair of an agent's symbolic knowledge to
do task planning in an incomplete or changing environmental model, but these
advances up to now have not been transferred to real physical robots. This
paper demonstrates how a physical robot can be capable of adapting its symbolic
knowledge of the environment, by using experiences in robot action execution to
drive knowledge refinement and hence to improve the success rate of the task
plans the robot creates. To implement more robust planning systems, we propose
a method for refining domain knowledge to improve the knowledge on which
intelligent robot behavior is based. This architecture has been implemented and
evaluated using a NAO robot. The refined knowledge leads to the future
synthesis of task plans which demonstrate decreasing rates of failure over time
as faulty knowledge is removed or adjusted.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:43:33 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Jazzaa",
"Hadeel",
""
],
[
"McCluskey",
"Thomas",
""
],
[
"Peebles",
"David",
""
]
] |
2504.14260 | Xiao Liu | Liu Xiao, Li Zhiyuan, Lin Yueyu | Cross-attention for State-based model RWKV-7 | null | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by/4.0/ | We introduce CrossWKV, a novel cross-attention mechanism for the state-based
RWKV-7 model, designed to enhance the expressive power of text-to-image
generation. Leveraging RWKV-7's linear-complexity Weighted Key-Value (WKV)
architecture, CrossWKV integrates text and image modalities in a single pass,
utilizing a generalized delta rule with vector-valued gating and low-rank
adaptations (LoRA) to achieve superior cross-modal alignment. Unlike
Transformer-based models, CrossWKV's non-diagonal, input-dependent transition
matrix enables it to represent complex functions beyond the $\mathrm{TC}^0$
complexity class, including all regular languages, as demonstrated by its
ability to perform state-tracking tasks like $S_5$ permutation modeling.
Evaluated within the Diffusion in RWKV-7 (DIR-7) on datasets such as LAION-5B
and ImageNet, CrossWKV achieves a Frechet Inception Distance (FID) of 2.88 and
a CLIP score of 0.33 on ImageNet 256x256, matching state-of-the-art performance
while offering robust generalization across diverse prompts. The model's
enhanced expressivity, combined with constant memory usage and linear scaling,
positions it as a powerful solution for advanced cross-modal tasks, with
potential applications in high-resolution generation and dynamic state
manipulation.Code at https://github.com/TorchRWKV/flash-linear-attention
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:47:51 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Xiao",
"Liu",
""
],
[
"Zhiyuan",
"Li",
""
],
[
"Yueyu",
"Lin",
""
]
] |
2504.14261 | Darius Guma | Herbert Batte and Darius Guma | On $k$-Pell numbers that are Palindromes formed by two distinct
Repdigits | 14 pages | null | null | null | math.NT | http://creativecommons.org/licenses/by/4.0/ | Let $k \ge 2$ and consider the sequence $\{P_n^{(k)}\}_{n \ge 2-k}$ of
$k$-generalized Pell numbers, which begins with the first $k$ terms as $0,
\ldots, 0, 0, 1$, and satisfies the recurrence relation $P_n^{(k)} =
2P_{n-1}^{(k)} + P_{n-2}^{(k)} + \cdots + P_{n-k}^{(k)}$ for all $n \ge 2$. In
this work, we identify all terms in the $k$-Pell sequence that can be expressed
as palindromes formed by concatenating two distinct repdigits.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 10:54:41 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Batte",
"Herbert",
""
],
[
"Guma",
"Darius",
""
]
] |
2504.14262 | Yoshinari Takeishi | Yoshinari Takeishi, Jun'ichi Takeuchi | Sparse Superposition Codes with Binomial Dictionary are
Capacity-Achieving with Maximum Likelihood Decoding | null | null | null | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is known that sparse superposition codes asymptotically achieve the
channel capacity over the additive white Gaussian noise channel with both
maximum likelihood decoding and efficient decoding (Joseph and Barron in 2012,
2014). Takeishi et al. (in 2014, 2019) demonstrated that these codes can also
asymptotically achieve the channel capacity with maximum likelihood decoding
when the dictionary is drawn from a Bernoulli distribution. In this paper, we
extend these results by showing that the dictionary distribution can be
naturally generalized to the binomial distribution.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:07:16 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Takeishi",
"Yoshinari",
""
],
[
"Takeuchi",
"Jun'ichi",
""
]
] |
2504.14263 | Alonso Beaumont | Alonso Beaumont | A uniform Tits alternative for endomorphisms of the projective line | null | null | null | null | math.NT math.DS math.GR | http://creativecommons.org/licenses/by/4.0/ | A recent article of J.P. Bell, K. Huang, W. Peng and T.J. Tucker establishes
an analog of the Tits alternative for semigroups of endomorphisms of the
projective line. The proof involves a ping-pong argument on arithmetic height
functions. Extending this method, we obtain a uniform version of the same
alternative. In particular, we show that semigroups of
$\mathrm{End}(\mathbb{P}^{1})$ of exponential growth are of uniform exponential
growth.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:09:44 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Beaumont",
"Alonso",
""
]
] |
2504.14264 | Juan Nathaniel | Juan Nathaniel and Pierre Gentine | Generative emulation of chaotic dynamics with coherent prior | 41 pages, 25 figures | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Data-driven emulation of nonlinear dynamics is challenging due to long-range
skill decay that often produces physically unrealistic outputs. Recent advances
in generative modeling aim to address these issues by providing uncertainty
quantification and correction. However, the quality of generated simulation
remains heavily dependent on the choice of conditioning priors. In this work,
we present an efficient generative framework for dynamics emulation, unifying
principles of turbulence with diffusion-based modeling: Cohesion. Specifically,
our method estimates large-scale coherent structure of the underlying dynamics
as guidance during the denoising process, where small-scale fluctuation in the
flow is then resolved. These coherent priors are efficiently approximated using
reduced-order models, such as deep Koopman operators, that allow for rapid
generation of long prior sequences while maintaining stability over extended
forecasting horizon. With this gain, we can reframe forecasting as trajectory
planning, a common task in reinforcement learning, where conditional denoising
is performed once over entire sequences, minimizing the computational cost of
autoregressive-based generative methods. Empirical evaluations on chaotic
systems of increasing complexity, including Kolmogorov flow, shallow water
equations, and subseasonal-to-seasonal climate dynamics, demonstrate Cohesion
superior long-range forecasting skill that can efficiently generate
physically-consistent simulations, even in the presence of partially-observed
guidance.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:14:40 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Nathaniel",
"Juan",
""
],
[
"Gentine",
"Pierre",
""
]
] |
2504.14265 | Benjamin Gillott | Benjamin Gillott | A Covering Pursuit Game | 18 pages | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the `Covering' pursuit game on a graph, a robber and a set of cops play
alternately, with the cops each moving to an adjacent vertex (or not moving)
and the robber moving to a vertex at distance at most 2 from his current
vertex. The aim of the cops is to ensure that, after every one of their turns,
there is a cop at the same vertex as the robber. How few cops are needed?
Our main aim in this paper is to consider this problem for the
two-dimensional grid $[n]^2$. Bollob\'{a}s and Leader asked if the number of
cops needed is $o(n^2)$. We answer this question by showing that $n^{1.999}$
cops suffice. We also consider some applications. In particular we study the
game `Catching a Fast Robber', concerning the number of cops needed to catch a
fast robber of speed $s$ on the two-dimensional grid $[n]^2$. We improve the
bounds proved by Balister, Bollob\'{a}s, Narayanan and Shaw for this game.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:21:29 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Gillott",
"Benjamin",
""
]
] |
2504.14266 | Anneswa Paul Ms | Anneswa Paul and Upendra Harbola | Light-matter interaction inside an optical cavity: A perspective | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Light-matter interaction inside an optical cavity and formation of
polaritonic states have gained interest in the past decades as it has direct
applications in many research fields. Different regimes of light-matter
coupling have been studied using different approximations, one of which is
rotating wave approximation(RWA), which is said to be valid for weak coupling
and near resonant regimes. In this study, we have categorized the light-matter
coupling into four regimes depending on the validity of the RWA as moderate,
$\lambda/\omega_c\leq 0.1$, strong, $0.1 \leq \lambda/\omega_c\leq 0.5$,
ultra-strong, $0.5 \leq\lambda/\omega_c\leq 1.0$ and deep-strong,
$\lambda/\omega_c\geq 1.0$ coupling, where $\lambda$ is the coupling strength
and $\omega_c$ is the cavity frequency. In experiments, vacuum Rabi-splitting
has been observed which is a clear indication of formation of polaritonic
states. It is a common misunderstanding that the cavity remains in vacuum state
when the coupled system is in the ground state. Here we show that upon
coupling, the cavity has non-zero excitation even in the ground state, which is
not captured by RWA. In fact, RWA breaks down completely to predict the ground
state properties as it fails to capture the interaction between the matter and
the cavity field.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:24:45 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Paul",
"Anneswa",
""
],
[
"Harbola",
"Upendra",
""
]
] |
2504.14267 | Xuanzhe Sun | Li Yu, Xuanzhe Sun, Wei Zhou, Moncef Gabbouj | Text-Audio-Visual-conditioned Diffusion Model for Video Saliency
Prediction | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video saliency prediction is crucial for downstream applications, such as
video compression and human-computer interaction. With the flourishing of
multimodal learning, researchers started to explore multimodal video saliency
prediction, including audio-visual and text-visual approaches. Auditory cues
guide the gaze of viewers to sound sources, while textual cues provide semantic
guidance for understanding video content. Integrating these complementary cues
can improve the accuracy of saliency prediction. Therefore, we attempt to
simultaneously analyze visual, auditory, and textual modalities in this paper,
and propose TAVDiff, a Text-Audio-Visual-conditioned Diffusion Model for video
saliency prediction. TAVDiff treats video saliency prediction as an image
generation task conditioned on textual, audio, and visual inputs, and predicts
saliency maps through stepwise denoising. To effectively utilize text, a large
multimodal model is used to generate textual descriptions for video frames and
introduce a saliency-oriented image-text response (SITR) mechanism to generate
image-text response maps. It is used as conditional information to guide the
model to localize the visual regions that are semantically related to the
textual description. Regarding the auditory modality, it is used as another
conditional information for directing the model to focus on salient regions
indicated by sounds. At the same time, since the diffusion transformer (DiT)
directly concatenates the conditional information with the timestep, which may
affect the estimation of the noise level. To achieve effective conditional
guidance, we propose Saliency-DiT, which decouples the conditional information
from the timestep. Experimental results show that TAVDiff outperforms existing
methods, improving 1.03\%, 2.35\%, 2.71\% and 0.33\% on SIM, CC, NSS and AUC-J
metrics, respectively.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:30:54 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Yu",
"Li",
""
],
[
"Sun",
"Xuanzhe",
""
],
[
"Zhou",
"Wei",
""
],
[
"Gabbouj",
"Moncef",
""
]
] |
2504.14268 | Xinye Chen | Xinye Chen | Mixed-Precision Conjugate Gradient Solvers with RL-Driven Precision
Tuning | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | This paper presents a novel reinforcement learning (RL) framework for
dynamically optimizing numerical precision in the preconditioned conjugate
gradient (CG) method. By modeling precision selection as a Markov Decision
Process (MDP), we employ Q-learning to adaptively assign precision levels to
key operations, striking an optimal balance between computational efficiency
and numerical accuracy, while ensuring stability through double-precision
scalar computations and residual computing. In practice, the algorithm is
trained on a set of data and subsequently performs inference for precision
selection on out-of-sample data, without requiring re-analysis or retraining
for new datasets. This enables the method to adapt seamlessly to new problem
instances without the computational overhead of recalibration. Our results
demonstrate the effectiveness of RL in enhancing solver's performance, marking
the first application of RL to mixed-precision numerical methods. The findings
highlight the approach's practical advantages, robustness, and scalability,
providing valuable insights into its integration with iterative solvers and
paving the way for AI-driven advancements in scientific computing.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:35:03 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Chen",
"Xinye",
""
]
] |
2504.14269 | Md. Humaun Kabir | Saif Bashar, Samia Nasir Nira, Shabbir Mahmood, Md. Humaun Kabir,
Sujit Roy, Iffat Farhana | Recognition of Frequencies of Short-Time SSVEP Signals Utilizing an
SSCCA-Based Spatio-Spectral Feature Fusion Framework | This paper is presented in the Undergraduate Conference on
Intelligent Computing and Systems (UCICS 2025) organised by Varendra
University, Rajshahi, Bangladesh, 2025 | null | null | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | A brain-computer interface (BCI) facilitates direct communication between the
brain and external equipment through EEG, which is preferred for its superior
temporal resolution. Among EEG techniques, the steady-state visual evoked
potential (SSVEP) is favored due to its robust signal-to-noise ratio, minimal
training demands, and elevated information transmission rate. Frequency
detection in SSVEP-based brain-computer interfaces commonly employs canonical
correlation analysis (CCA). SSCCA (spatio-spectral canonical correlation
analysis) augments CCA by refining spatial filtering. This paper presents a
multistage feature fusion methodology for short-duration SSVEP frequency
identification, employing SSCCA with template signals derived via leave-one-out
cross-validation (LOOCV). A filterbank generates bandpass filters for stimulus
frequencies and their harmonics, whereas SSCCA calculates correlation
coefficients between subbands and templates. Two phases of non-linear weighting
amalgamate these coefficients to discern the target stimulus. This multistage
methodology surpasses traditional techniques, attaining a accuracy of 94.5%.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:42:06 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Bashar",
"Saif",
""
],
[
"Nira",
"Samia Nasir",
""
],
[
"Mahmood",
"Shabbir",
""
],
[
"Kabir",
"Md. Humaun",
""
],
[
"Roy",
"Sujit",
""
],
[
"Farhana",
"Iffat",
""
]
] |
2504.14271 | Kai Zhang | Rui.Guo, Kai Zhang, Nicholas X. Fang | Wave characteristics and anisotropic homogenization theory of soft
matters layered structure | 10 pages, 4 figures | null | null | null | cond-mat.soft | http://creativecommons.org/licenses/by/4.0/ | We investigate in this work the wave characteristics and homogenization
theory of soft matter layered structure in the limit of low-frequency P-wave.
Using the method of potentials, we derive closed-form dispersion relationship
and identify three distinct modes of the soft matter layered structure:
quasistatic mode, resonance mode, and slip mode. These modes differ based on
their equivalent interface conditions: a continuous interface for quasistatic
mode, a spring-like interface for resonance mode, and a slip like interface for
slip mode. Additionally, we propose a simplified model capturing P-wave wave
characteristics in the S-wave high-frequency regime. Our findings unify
wave-structure relationships across solid, liquid, and soft matter composites,
offering a predictive framework for engineering metamaterials with programmable
wave responses. This study offers new insight on the fundamental understanding
of layered media and providing direct design principles for applications in
acoustic cloaking, vibration damping, and biomedical imaging.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:47:31 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Guo",
"Rui.",
""
],
[
"Zhang",
"Kai",
""
],
[
"Fang",
"Nicholas X.",
""
]
] |
2504.14272 | Daeho Um | Yunha Yeo, Daeho Um | Can AI Recognize the Style of Art? Analyzing Aesthetics through the Lens
of Style Transfer | Accepted to ISEA 2025 | null | null | null | cs.GR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This study investigates how artificial intelligence (AI) recognizes style
through style transfer-an AI technique that generates a new image by applying
the style of one image to another. Despite the considerable interest that style
transfer has garnered among researchers, most efforts have focused on enhancing
the quality of output images through advanced AI algorithms. In this paper, we
approach style transfer from an aesthetic perspective, thereby bridging AI
techniques and aesthetics. We analyze two style transfer algorithms: one based
on convolutional neural networks (CNNs) and the other utilizing recent
Transformer models. By comparing the images produced by each, we explore the
elements that constitute the style of artworks through an aesthetic analysis of
the style transfer results. We then elucidate the limitations of current style
transfer techniques. Based on these limitations, we propose potential
directions for future research on style transfer techniques.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:51:43 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Yeo",
"Yunha",
""
],
[
"Um",
"Daeho",
""
]
] |
2504.14274 | Zhengxi Lu | Zhengxi Lu, Shizhuo Cheng, Yuru Jiang, Yan Zhang, Min Zhang | ProtPainter: Draw or Drag Protein via Topology-guided Diffusion | Published as a conference paper at ICLR 2025 | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent advances in protein backbone generation have achieved promising
results under structural, functional, or physical constraints. However,
existing methods lack the flexibility for precise topology control, limiting
navigation of the backbone space. We present ProtPainter, a diffusion-based
approach for generating protein backbones conditioned on 3D curves. ProtPainter
follows a two-stage process: curve-based sketching and sketch-guided backbone
generation. For the first stage, we propose CurveEncoder, which predicts
secondary structure annotations from a curve to parametrize sketch generation.
For the second stage, the sketch guides the generative process in Denoising
Diffusion Probabilistic Modeling (DDPM) to generate backbones. During this
process, we further introduce a fusion scheduling scheme, Helix-Gating, to
control the scaling factors. To evaluate, we propose the first benchmark for
topology-conditioned protein generation, introducing Protein Restoration Task
and a new metric, self-consistency Topology Fitness (scTF). Experiments
demonstrate ProtPainter's ability to generate topology-fit (scTF > 0.8) and
designable (scTM > 0.5) backbones, with drawing and dragging tasks showcasing
its flexibility and versatility.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 11:59:48 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Lu",
"Zhengxi",
""
],
[
"Cheng",
"Shizhuo",
""
],
[
"Jiang",
"Yuru",
""
],
[
"Zhang",
"Yan",
""
],
[
"Zhang",
"Min",
""
]
] |
2504.14275 | Lenka Ptackova | Lenka Ptackova | Leibniz rule for wedge product in discrete exterior calculus on general
polygonal meshes | null | null | null | null | math.AT cs.CG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Discrete exterior calculus offers a coordinate-free discretization of
exterior calculus especially suited for computations on curved spaces. In this
work, we present a wedge product on 2-dimensional pseudomanifolds, whose faces
are any polygons. We prove that this polygonal wedge product is compatible with
the discrete exterior derivative in the sense that it satisfies the Leibniz
product rule. We thus extend previously studied discretizations of wedge
products from simplicial or quadrilateral meshes to general polygonal surface
meshes. We also prove that our discrete wedge product corresponds to a cup
product of cochains on 2-pseudomanifolds.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:00:05 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ptackova",
"Lenka",
""
]
] |
2504.14276 | Sasmita Mishra | Vivek Banerjee, Sasmita Mishra | Magnetic moment of neutrinos in a left-right symmetric model and
Interplay of type-I and type-II seesaw | null | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | In left-right symmetric models, the Majorana coupling matrix, $f$, and hence
the right-handed neutrino (RHN) mass matrix, admits eight solutions assuming
the form of Dirac coupling matrix is known. Additionally, the coupling matrix
depends on the parity-breaking scale, $v_R$, as a new physics scale. RHNs being
Majorana in nature can possess a transition magnetic moment (TMM). Neutrino
magnetic moments are inherently related to neutrino masses, as neutrino masses
imply neutrino magnetic moments. We study, along with small neutrino TMM, the
heavy RHN transition magnetic moment contributions to muon $g-2$, $(g-2)_\mu$
anomaly for all eight solutions of $f$. We find, of the eight solutions, only
two solutions of $f$ matrix contribute to the $(g-2)_\mu$ in the experimental
predicted range. The range of $v_R$ in these cases is found to be $3.4 \times
10^3 - 1.5 \times 10^4$ GeV. For a complementary check, we also study TMM
induced neutrinoless double beta decay ($0\nu\beta\beta$), for the same set of
choice of parameters. While certain parameter choices allow RHNs to explain the
$(g-2)_\mu$ anomaly, the same configurations lead to an extremely long
half-life for $0\nu\beta\beta$ decay, well beyond experimental reach. Even
under extreme magnetic field enhancements, the half-life decreases only
marginally, reinforcing the dominance of weak interaction vertices over TMM
contributions in $0\nu\beta\beta$ decay.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:10:49 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Banerjee",
"Vivek",
""
],
[
"Mishra",
"Sasmita",
""
]
] |
2504.14277 | Chengxi Li | Chengxi Li and Kai Rajala | Uniformization of cofat domains on metric two-spheres | null | null | null | null | math.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We extend \emph{Schramm's cofat uniformization theorem} to cofat domains on
upper Ahlfors 2-regular metric two-spheres $X$. Specifically, we show that if
$\Omega \subset X$ is a cofat domain, then there exists a
$\frac{\pi}{2}$-quasiconformal homeomorphism $f: \Omega \to D$ onto a circle
domain $D \subset \mathbb{S}^2$. Moreover, $f$ preserves the point-components
and non-trivial complementary components. We also construct examples which show
that the above conclusions are not true for countably connected
$\ell^{\alpha}$-subdomains of $\mathbb{S}^2$.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:12:26 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Li",
"Chengxi",
""
],
[
"Rajala",
"Kai",
""
]
] |
2504.14278 | Yuke Hou | Shang Zhang, Yuke Hou, Guoqiang Gong, Ruoyan Xiong, and Yue Zhang | RAMCT: Novel Region-adaptive Multi-channel Tracker with Iterative
Tikhonov Regularization for Thermal Infrared Tracking | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Correlation filter (CF)-based trackers have gained significant attention for
their computational efficiency in thermal infrared (TIR) target tracking.
However, ex-isting methods struggle with challenges such as low-resolution
imagery, occlu-sion, background clutter, and target deformation, which severely
impact tracking performance. To overcome these limitations, we propose RAMCT, a
region-adaptive sparse correlation filter tracker that integrates multi-channel
feature opti-mization with an adaptive regularization strategy. Firstly, we
refine the CF learn-ing process by introducing a spatially adaptive binary
mask, which enforces spar-sity in the target region while dynamically
suppressing background interference. Secondly, we introduce generalized
singular value decomposition (GSVD) and propose a novel GSVD-based
region-adaptive iterative Tikhonov regularization method. This enables flexible
and robust optimization across multiple feature channels, improving resilience
to occlusion and background variations. Thirdly, we propose an online
optimization strategy with dynamic discrepancy-based pa-rameter adjustment.
This mechanism facilitates real time adaptation to target and background
variations, thereby improving tracking accuracy and robustness. Ex-tensive
experiments on LSOTB-TIR, PTB-TIR, VOT-TIR2015, and VOT-TIR2017 benchmarks
demonstrate that RAMCT outperforms other state-of-the-art trackers in terms of
accuracy and robustness.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:18:36 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Shang",
""
],
[
"Hou",
"Yuke",
""
],
[
"Gong",
"Guoqiang",
""
],
[
"Xiong",
"Ruoyan",
""
],
[
"Zhang",
"Yue",
""
]
] |
2504.14279 | Majid Zamani | Xiaoyu Jiang, Tao Fang and Majid Zamani | A Real-time and Hardware Efficient Artfecat-free Spike Sorting Using
Deep Spike Detection | null | null | null | null | eess.SP | http://creativecommons.org/licenses/by/4.0/ | Spike sorting is a valuable tool in understanding brain regions. It assigns
detected spike waveforms to their origins, helping to research the mechanism of
the human brain and the development of implantable brain-machine interfaces
(iBMIs). The presence of noise and artefacts will adversely affect the efficacy
of spike sorting. This paper proposes a framework for low-cost and real-time
implementation of deep spike detection, which consists of two one-dimensional
(1-D) convolutional neural network (CNN) model for channel selection and
artefact removal. The framework utilizes simulation and hardware layers, and it
applies several low-power techniques to optimise the implementation cost of a
1-D CNN model. A compact CNN model with 210 bytes memory size is achieved using
structured pruning, network projection and quantization in the simulation
layer. The hardware layer also accommodates various techniques including a
customized multiply-accumulate (MAC) engine, novel fused layers in the
convolution pipeline and proposing flexible resource allocation for a
power-efficient and low-delay design. The optimized 1-D CNN significantly
decreases both computational complexity and model size, with only a minimal
reduction in accuracy. Classification of 1-D CNN on the Cyclone V 5CSEMA5F31C6
FPGA evaluation platform is accomplished in just 16.8 microseconds at a
frequency of 2.5 MHz. The FPGA prototype achieves an accuracy rate of 97.14% on
a standard dataset and operates with a power consumption of 2.67mW from a
supply voltage of 1.1 volts. An accuracy of 95.05% is achieved with a power of
5.6mW when deep spike detection is implemented using two optimized 1-D CNNs on
an FPGA board.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:19:09 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Jiang",
"Xiaoyu",
""
],
[
"Fang",
"Tao",
""
],
[
"Zamani",
"Majid",
""
]
] |
2504.14280 | Jindong Li | Jindong Li, Yongguang Li, Yali Fu, Jiahong Liu, Yixin Liu, Menglin
Yang, Irwin King | CLIP-Powered Domain Generalization and Domain Adaptation: A
Comprehensive Survey | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | As machine learning evolves, domain generalization (DG) and domain adaptation
(DA) have become crucial for enhancing model robustness across diverse
environments. Contrastive Language-Image Pretraining (CLIP) plays a significant
role in these tasks, offering powerful zero-shot capabilities that allow models
to perform effectively in unseen domains. However, there remains a significant
gap in the literature, as no comprehensive survey currently exists that
systematically explores the applications of CLIP in DG and DA, highlighting the
necessity for this review. This survey presents a comprehensive review of
CLIP's applications in DG and DA. In DG, we categorize methods into optimizing
prompt learning for task alignment and leveraging CLIP as a backbone for
effective feature extraction, both enhancing model adaptability. For DA, we
examine both source-available methods utilizing labeled source data and
source-free approaches primarily based on target domain data, emphasizing
knowledge transfer mechanisms and strategies for improved performance across
diverse contexts. Key challenges, including overfitting, domain diversity, and
computational efficiency, are addressed, alongside future research
opportunities to advance robustness and efficiency in practical applications.
By synthesizing existing literature and pinpointing critical gaps, this survey
provides valuable insights for researchers and practitioners, proposing
directions for effectively leveraging CLIP to enhance methodologies in domain
generalization and adaptation. Ultimately, this work aims to foster innovation
and collaboration in the quest for more resilient machine learning models that
can perform reliably across diverse real-world scenarios. A more up-to-date
version of the papers is maintained at:
https://github.com/jindongli-Ai/Survey_on_CLIP-Powered_Domain_Generalization_and_Adaptation.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:27:24 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Li",
"Jindong",
""
],
[
"Li",
"Yongguang",
""
],
[
"Fu",
"Yali",
""
],
[
"Liu",
"Jiahong",
""
],
[
"Liu",
"Yixin",
""
],
[
"Yang",
"Menglin",
""
],
[
"King",
"Irwin",
""
]
] |
2504.14281 | Ly V. Nguyen | Ly V. Nguyen, Rang Liu, Nhan Thanh Nguyen, Markku Juntti, Bj\"orn
Ottersten, and A. Lee Swindlehurst | Exploiting Symmetric Non-Convexity for Multi-Objective Symbol-Level DFRC
Signal Design | 13 pages, 8 figures | null | null | null | eess.SP | http://creativecommons.org/licenses/by/4.0/ | Symbol-level precoding (SLP) is a promising solution for addressing the
inherent interference problem in dual-functional radar-communication (DFRC)
signal designs. This paper considers an SLP-DFRC signal design problem which
optimizes the radar performance under communication performance constraints. We
show that a common phase modulation applied to the transmit signals from an
antenna array does not affect the performance of different radar sensing
metrics, including beampattern similarity, signal-to-interference-plus-noise
ratio (SINR), and Cram\'er-Rao lower bound (CRLB). We refer to this as
symmetric-rotation invariance, upon which we develop low-complexity yet
efficient DFRC signal design algorithms. More specifically, we propose a
symmetric non-convexity (SNC)-based DFRC algorithm that relies on the
non-convexity of the radar sensing metrics to identify a set of radar-only
solutions. Based on these solutions, we further exploit the symmetry property
of the radar sensing metrics to efficiently design the DFRC signal. We show
that the proposed SNC-based algorithm is versatile in the sense that it can be
applied to the DFRC signal optimization of all three sensing metrics mentioned
above (beampattern, SINR, and CRLB). In addition, since the radar sensing
metrics are independent of the communication channel and data symbols, the set
of radar-only solutions can be constructed offline, thereby reducing the
computational complexity. We also develop an accelerated SNC-based algorithm
that further reduces the complexity. Finally, we numerically demonstrate the
superiority of the proposed algorithms compared to existing methods in terms of
sensing and communication performance as well as computational requirements.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:29:11 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Nguyen",
"Ly V.",
""
],
[
"Liu",
"Rang",
""
],
[
"Nguyen",
"Nhan Thanh",
""
],
[
"Juntti",
"Markku",
""
],
[
"Ottersten",
"Björn",
""
],
[
"Swindlehurst",
"A. Lee",
""
]
] |
2504.14282 | Ze Zhao | Ze Zhao, Bin Lu, Xiaoying Gan, Gu Tang, Luoyi Fu, Xinbing Wang | CHAINSFORMER: Numerical Reasoning on Knowledge Graphs from a Chain
Perspective | Accepted to ICDE 2025 | null | null | null | cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reasoning over Knowledge Graphs (KGs) plays a pivotal role in knowledge graph
completion or question answering systems, providing richer and more accurate
triples and attributes. As numerical attributes become increasingly essential
in characterizing entities and relations in KGs, the ability to reason over
these attributes has gained significant importance. Existing graph-based
methods such as Graph Neural Networks (GNNs) and Knowledge Graph Embeddings
(KGEs), primarily focus on aggregating homogeneous local neighbors and
implicitly embedding diverse triples. However, these approaches often fail to
fully leverage the potential of logical paths within the graph, limiting their
effectiveness in exploiting the reasoning process. To address these
limitations, we propose ChainsFormer, a novel chain-based framework designed to
support numerical reasoning. Chainsformer not only explicitly constructs
logical chains but also expands the reasoning depth to multiple hops.
Specially, we introduces Relation-Attribute Chains (RA-Chains), a specialized
logic chain, to model sequential reasoning patterns. ChainsFormer captures the
step-by-step nature of multi-hop reasoning along RA-Chains by employing
sequential in-context learning. To mitigate the impact of noisy chains, we
propose a hyperbolic affinity scoring mechanism that selects relevant logic
chains in a variable-resolution space. Furthermore, ChainsFormer incorporates
an attention-based numerical reasoner to identify critical reasoning paths,
enhancing both reasoning accuracy and transparency. Experimental results
demonstrate that ChainsFormer significantly outperforms state-of-the-art
methods, achieving up to a 20.0% improvement in performance. The
implementations are available at
https://github.com/zhaodazhuang2333/ChainsFormer.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:47:44 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhao",
"Ze",
""
],
[
"Lu",
"Bin",
""
],
[
"Gan",
"Xiaoying",
""
],
[
"Tang",
"Gu",
""
],
[
"Fu",
"Luoyi",
""
],
[
"Wang",
"Xinbing",
""
]
] |
2504.14283 | Quang Loc Le | James Brotherston, Quang Loc Le, Gauri Desai and Yukihiro Oda | Cyclic Proofs in Hoare Logic and its Reverse | null | null | null | null | cs.LO cs.PL | http://creativecommons.org/licenses/by/4.0/ | We examine the relationships between axiomatic and cyclic proof systems for
the partial and total versions of Hoare logic and those of its dual, known as
reverse Hoare logic (or sometimes incorrectness logic). In the axiomatic proof
systems for these logics, the proof rules for looping constructs involve an
explicit loop invariant, which in the case of the total versions additionally
require a well-founded termination measure. In the cyclic systems, these are
replaced by rules that simply unroll the loops, together with a principle
allowing the formation of cycles in the proof, subject to a global soundness
condition that ensures the well-foundedness of the circular reasoning.
Interestingly, the cyclic soundness conditions for partial Hoare logic and its
reverse are similar and essentially coinductive in character, while those for
the total versions are also similar and essentially inductive. We show that
these cyclic systems are sound, by direct argument, and relatively complete, by
translation from axiomatic to cyclic proofs.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 12:51:18 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Brotherston",
"James",
""
],
[
"Le",
"Quang Loc",
""
],
[
"Desai",
"Gauri",
""
],
[
"Oda",
"Yukihiro",
""
]
] |
2504.14284 | Satoshi Fujii | Satoshi Fujii | On automorphisms of some semidirect product groups and ranks of Iwasawa
modules | null | null | null | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $p$ be an odd prime number and $k$ an imaginary quadratic field in which
$p$ does not split. Based on their heuristic, Kundu and Washington posed a
question which asks whether $\lambda$- and $\mu$-invariant of the
anti-cyclotomic ${\Bbb Z}_p$-extension $k_{\infty}^a$ of $k$ are always
trivial. Also, if $k_{\infty}^a/k$ is totally ramified, for $n\geq 1$, they
showed that the $p$-part of the ideal class group of the $n$th layer of the
anti-cyclotomic ${\Bbb Z}_p$-extension of $k$ is not cyclic. In this article,
inspired by their paper, we study anti-cyclotomic like ${\Bbb Z}_p$-extensions,
extending both the above question and Kundu-Washington's result. We show that
the values of $\lambda$ of certain anti-cyclotomic like ${\Bbb Z}_p$-extensions
are always even. We also show the $p$-part of the ideal class groups of certain
anti-cyclotomic like ${\Bbb Z}_p$-extensions of CM-fields are always not
cyclic.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:00:34 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Fujii",
"Satoshi",
""
]
] |
2504.14285 | Dominik Krasula | Dominik Krasula | Formal matrix representations of pseudo-Frobenius and Frobenius rings | null | null | null | null | math.RA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Rings with Nakayama permutations, pseudo-Frobenius and Frobenius rings in
particular, are studied by applying the general theory of formal matrix rings
to their Peirce decompositions. A combinatorial criterion is given to decide
whether a formal matrix ring with local rings on the diagonal has a prescribed
Nakayama permutation.
It is shown that a pseudo-Frobenius ring R can be represented as a block
matrix ring, where the blocks on the diagonal are pseudo-Frobenius rings
corresponding to cycles in the Nakayama permutation of R. All possible supports
of such blocks are characterised. In the finite case, their local corner rings
are shown to be isomorphic. We characterise local corners of quasi-Frobenius
rings as a subclass of rings with a Morita self-duality. The duality contexts
between these corners then appear on the shifted diagonal of their formal
matrix representations.
Using the combinatorial criterion, we give, under mild assumptions, a method
of how to glue two rings with a Nakayama permutation. It is then used to
construct an indecomposable Frobenius ring with two simple modules whose rings
of endomorphisms are not isomorphic.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:05:43 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Krasula",
"Dominik",
""
]
] |
2504.14287 | Demetris Paschalides | Demetris Paschalides, George Pallis, Marios D. Dikaiakos | Probing the Subtle Ideological Manipulation of Large Language Models | null | null | null | null | cs.CL cs.CY | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have transformed natural language processing,
but concerns have emerged about their susceptibility to ideological
manipulation, particularly in politically sensitive areas. Prior work has
focused on binary Left-Right LLM biases, using explicit prompts and fine-tuning
on political QA datasets. In this work, we move beyond this binary approach to
explore the extent to which LLMs can be influenced across a spectrum of
political ideologies, from Progressive-Left to Conservative-Right. We introduce
a novel multi-task dataset designed to reflect diverse ideological positions
through tasks such as ideological QA, statement ranking, manifesto cloze
completion, and Congress bill comprehension. By fine-tuning three LLMs-Phi-2,
Mistral, and Llama-3-on this dataset, we evaluate their capacity to adopt and
express these nuanced ideologies. Our findings indicate that fine-tuning
significantly enhances nuanced ideological alignment, while explicit prompts
provide only minor refinements. This highlights the models' susceptibility to
subtle ideological manipulation, suggesting a need for more robust safeguards
to mitigate these risks.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:11:50 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Paschalides",
"Demetris",
""
],
[
"Pallis",
"George",
""
],
[
"Dikaiakos",
"Marios D.",
""
]
] |
2504.14288 | Hanxiao Wang | Qi L\"u, Bowen Ma and Hanxiao Wang | Forward-Backward Stochastic Linear-Quadratic Optimal Controls:
Equilibrium Strategies and Non-Symmetric Riccati Equations | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | Linear-quadratic optimal control problem for systems governed by
forward-backward stochastic differential equations has been extensively studied
over the past three decades. Recent research has revealed that for
forward-backward control systems, the corresponding optimal control problem is
inherently time-inconsistent. Consequently, the optimal controls derived in
existing literature represent pre-committed solutions rather than dynamically
consistent strategies. In this paper, we shift focus from pre-committed
solutions to addressing the time-inconsistency issue directly, adopting a
dynamic game-theoretic approach to derive equilibrium strategies. Owing to the
forward-backward structure, the associated equilibrium Riccati equation (ERE)
constitutes a coupled system of matrix-valued, non-local ordinary differential
equations with a non-symmetric structure. This non-symmetry introduces
fundamental challenges in establishing the solvability of the EREs. We overcome
the difficulty by establishing a priori estimates for a combination of the
solutions to EREs, which, interestingly, is a representation of the equilibrium
value function.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:13:13 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Lü",
"Qi",
""
],
[
"Ma",
"Bowen",
""
],
[
"Wang",
"Hanxiao",
""
]
] |
2504.14289 | Yujie Cui | Shang Zhang, Yujie Cui, Ruoyan Xiong, and Huanbin Zhang | ISTD-YOLO: A Multi-Scale Lightweight High-Performance Infrared Small
Target Detection Algorithm | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Aiming at the detection difficulties of infrared images such as complex
background, low signal-to-noise ratio, small target size and weak brightness, a
lightweight infrared small target detection algorithm ISTD-YOLO based on
improved YOLOv7 was proposed. Firstly, the YOLOv7 network structure was
lightweight reconstructed, and a three-scale lightweight network architecture
was designed. Then, the ELAN-W module of the model neck network is replaced by
VoV-GSCSP to reduce the computational cost and the complexity of the network
structure. Secondly, a parameter-free attention mechanism was introduced into
the neck network to enhance the relevance of local con-text information.
Finally, the Normalized Wasserstein Distance (NWD) was used to optimize the
commonly used IoU index to enhance the localization and detection accuracy of
small targets. Experimental results show that compared with YOLOv7 and the
current mainstream algorithms, ISTD-YOLO can effectively improve the detection
effect, and all indicators are effectively improved, which can achieve
high-quality detection of infrared small targets.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:19:54 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Zhang",
"Shang",
""
],
[
"Cui",
"Yujie",
""
],
[
"Xiong",
"Ruoyan",
""
],
[
"Zhang",
"Huanbin",
""
]
] |
2504.14290 | Yao Huang | Shouwei Ruan, Zhenyu Wu, Yao Huang, Ruochen Zhang, Yitong Sun, Caixin
Kang, Xingxing Wei | Towards NSFW-Free Text-to-Image Generation via Safety-Constraint Direct
Preference Optimization | 10 pages, 6 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Ensuring the safety of generated content remains a fundamental challenge for
Text-to-Image (T2I) generation. Existing studies either fail to guarantee
complete safety under potentially harmful concepts or struggle to balance
safety with generation quality. To address these issues, we propose
Safety-Constrained Direct Preference Optimization (SC-DPO), a novel framework
for safety alignment in T2I models. SC-DPO integrates safety constraints into
the general human preference calibration, aiming to maximize the likelihood of
generating human-preferred samples while minimizing the safety cost of the
generated outputs. In SC-DPO, we introduce a safety cost model to accurately
quantify harmful levels for images, and train it effectively using the proposed
contrastive learning and cost anchoring objectives. To apply SC-DPO for
effective T2I safety alignment, we constructed SCP-10K, a safety-constrained
preference dataset containing rich harmful concepts, which blends
safety-constrained preference pairs under both harmful and clean instructions,
further mitigating the trade-off between safety and sample quality.
Additionally, we propose a Dynamic Focusing Mechanism (DFM) for SC-DPO,
promoting the model's learning of difficult preference pair samples. Extensive
experiments demonstrate that SC-DPO outperforms existing methods, effectively
defending against various NSFW content while maintaining optimal sample quality
and human preference alignment. Additionally, SC-DPO exhibits resilience
against adversarial prompts designed to generate harmful content.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:26:46 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Ruan",
"Shouwei",
""
],
[
"Wu",
"Zhenyu",
""
],
[
"Huang",
"Yao",
""
],
[
"Zhang",
"Ruochen",
""
],
[
"Sun",
"Yitong",
""
],
[
"Kang",
"Caixin",
""
],
[
"Wei",
"Xingxing",
""
]
] |
2504.14292 | Ari-Pekka Perkki\"o Mr. | Jean-Philippe Chancelier, Michel De Lara, Fran\c{c}ois Pacaud, Teemu
Pennanen, Ari-Pekka Perkki\"o | Optimal Operation and Valuation of Electricity Storages | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | This paper applies computational techniques of convex stochastic optimization
to optimal operation and valuation of electricity storages in the face of
uncertain electricity prices. Our approach is applicable to various
specifications of storages, and it allows for e.g.\ hard constraints on storage
capacity and charging speed. Our valuations are based on the indifference
pricing principle, which builds on optimal trading strategies and calibrates to
the user's initial position, market views and risk preferences. We illustrate
the effects of storage capacity and charging speed by numerically computing the
valuations using stochastic dual dynamic programming.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:35:33 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Chancelier",
"Jean-Philippe",
""
],
[
"De Lara",
"Michel",
""
],
[
"Pacaud",
"François",
""
],
[
"Pennanen",
"Teemu",
""
],
[
"Perkkiö",
"Ari-Pekka",
""
]
] |
2504.14293 | Leila Schneps | Leila Schneps | The double shuffle Lie algebra injects into the Kashiwara-Vergne Lie
algebra | null | null | null | null | math.RA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In this article we prove that there exists an injective Lie morphism from the
double shuffle Lie algebra ${\frak{ds}}$ into the Kashiwara-Vergne Lie algebra
${\frak{krv}}$, forming a commutative triangle with the known Lie injections of
the Grothendieck-Teichm\"uller Lie algebra ${\frak{grt}}\rightarrow
{\frak{ds}}$ and ${\frak{grt}}\rightarrow{\frak{krv}}$.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:39:05 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Schneps",
"Leila",
""
]
] |
2504.14294 | Pourya Shamsolmoali | Pourya Shamsolmoali, Masoumeh Zareapoor, Huiyu Zhou, Michael Felsberg,
Dacheng Tao, Xuelong Li | From Missing Pieces to Masterpieces: Image Completion with
Context-Adaptive Diffusion | Accepted in TPAMI | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Image completion is a challenging task, particularly when ensuring that
generated content seamlessly integrates with existing parts of an image. While
recent diffusion models have shown promise, they often struggle with
maintaining coherence between known and unknown (missing) regions. This issue
arises from the lack of explicit spatial and semantic alignment during the
diffusion process, resulting in content that does not smoothly integrate with
the original image. Additionally, diffusion models typically rely on global
learned distributions rather than localized features, leading to
inconsistencies between the generated and existing image parts. In this work,
we propose ConFill, a novel framework that introduces a Context-Adaptive
Discrepancy (CAD) model to ensure that intermediate distributions of known and
unknown regions are closely aligned throughout the diffusion process. By
incorporating CAD, our model progressively reduces discrepancies between
generated and original images at each diffusion step, leading to contextually
aligned completion. Moreover, ConFill uses a new Dynamic Sampling mechanism
that adaptively increases the sampling rate in regions with high reconstruction
complexity. This approach enables precise adjustments, enhancing detail and
integration in restored areas. Extensive experiments demonstrate that ConFill
outperforms current methods, setting a new benchmark in image completion.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:40:46 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Shamsolmoali",
"Pourya",
""
],
[
"Zareapoor",
"Masoumeh",
""
],
[
"Zhou",
"Huiyu",
""
],
[
"Felsberg",
"Michael",
""
],
[
"Tao",
"Dacheng",
""
],
[
"Li",
"Xuelong",
""
]
] |
2504.14295 | Bo-Qiang Ma | Hanlin Song, Bo-Qiang Ma | Examining Lorentz invariance violation with three remarkable GRB photons | 10 pages, 4 figures, final version as publication | Phys.Dark Univ.47(2025)101808 | 10.1016/j.dark.2025.101808 | null | astro-ph.HE gr-qc hep-ph | http://creativecommons.org/licenses/by/4.0/ | Lorentz invariance violation in photons can be quantified by measuring the
difference in arrival times between high- and low-energy photons originating
from gamma-ray bursts (GRBs). When analyzing data, it is crucial to consider
the inherent time delay in the emission of these photons at the source of the
GRB. In a recent study, three distinct models were evaluated to explain the
intrinsic emission times of high-energy photons by analyzing 14 multi-GeV
photon events detected from 8 GRBs using the Fermi Gamma-ray Space Telescope
(FGST). In this study, we examine three remarkable GRB photons recorded by
different observatories: the 99.3~GeV photon from GRB 221009A observed by FGST,
the 1.07~TeV photon from GRB 190114C detected by the Major Atmospheric Gamma
Imaging Cherenkov (MAGIC) telescope, and the 12.2~TeV photon from GRB 221009A
observed by the Large High Altitude Air-shower Observatory (LHAASO). Our
analysis indicates that the newly proposed model with a linear relationship
between photon energy and intrinsic emission time can offer a consistent
framework to explain the behavior of all three exceptional photons with a
Lorentz violation scale $E_{\rm LV}\sim 3\times 10^{17}$~GeV.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:45:50 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Song",
"Hanlin",
""
],
[
"Ma",
"Bo-Qiang",
""
]
] |
2504.14296 | Haiyan Wang | Haiyan Wang | Analysis of Discrete Stochastic Population Models with Normal
Distribution | null | null | null | null | math.PR q-bio.PE | http://creativecommons.org/licenses/by/4.0/ | This paper analyzes a stochastic logistic difference equation under the
assumption that the population distribution follows a normal distribution. Our
focus is on the mathematical relationship between the average growth rate and a
newly introduced concept, the uniform structural growth rate, which captures
how growth is influenced by the internal distributional structure of the
population. We derive explicit relationships linking the uniform structural
growth rate to the parameters of the normal distribution and the variance of a
small stochastic perturbation. The analysis reveals the existence of two
distinct branches of the uniform structural growth rate, corresponding to
alternative population states characterized by higher and lower growth rates.
This duality provides deeper insights into the dynamics of population growth
under stochastic influences. A sufficient condition for the existence of two
uniform structural growth rates is established and rigorously proved,
demonstrating that there exist infeasible intervals where no uniform structural
growth rate can be defined. We also explore the biological significance of
these findings, emphasizing the role of stochastic perturbations and the
distribution in shaping population dynamics.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:45:51 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Wang",
"Haiyan",
""
]
] |
2504.14297 | Tom\'a\v{s} Roub\'i\v{c}ek | Tom\'a\v{s} Roub\'i\v{c}ek | Time discretization in convected linearized thermo-visco-elastodynamics
at large displacements | null | null | null | null | math.NA cs.NA math.AP | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The fully-implicit time discretization (i.e. the backward Euler formula) is
applied to compressible nonlinear dynamical models of thermo-viscoelastic
solids in the Eulerian description, i.e. in the actual deforming configuration,
formulated in terms of rates. The Kelvin-Voigt rheology or also, in the
deviatoric part, the Jeffreys rheology (covering creep or plasticity) are
considered, using the additive Green-Naghdi's decomposition of total strain
into the elastic and the inelastic strains formulated in terms of (objective)
rates exploiting the Zaremba-Jaumann time derivative. A linearized convective
model at large displacements is considered, focusing on the case where the
internal energy additively splits the (convex) mechanical and the thermal
parts. The time-discrete suitably regularized scheme is devised. The numerical
stability and, considering the multipolar 2nd-grade viscosity, also convergence
towards weak solutions are proved, exploiting the convexity of the kinetic
energy when written in terms of linear momentum instead of velocity and
estimating the temperature gradient from the entropy-like inequality.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:48:55 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Roubíček",
"Tomáš",
""
]
] |
2504.14298 | Zhongsheng Fang | Xiucheng Wang, Zhongsheng Fang and Nan Cheng | RadioDiff-Inverse: Diffusion Enhanced Bayesian Inverse Estimation for
ISAC Radio Map Construction | 12 pages, 7 figures | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Radio maps (RMs) are essential for environment-aware communication and
sensing, providing location-specific wireless channel information. Existing RM
construction methods often rely on precise environmental data and base station
(BS) locations, which are not always available in dynamic or privacy-sensitive
environments. While sparse measurement techniques reduce data collection, the
impact of noise in sparse data on RM accuracy is not well understood. This
paper addresses these challenges by formulating RM construction as a Bayesian
inverse problem under coarse environmental knowledge and noisy sparse
measurements. Although maximum a posteriori (MAP) filtering offers an optimal
solution, it requires a precise prior distribution of the RM, which is
typically unavailable. To solve this, we propose RadioDiff-Inverse, a
diffusion-enhanced Bayesian inverse estimation framework that uses an
unconditional generative diffusion model to learn the RM prior. This approach
not only reconstructs the spatial distribution of wireless channel features but
also enables environmental structure perception, such as building outlines, and
location of BS just relay on pathloss, through integrated sensing and
communication (ISAC). Remarkably, RadioDiff-Inverse is training-free,
leveraging a pre-trained model from Imagenet without task-specific fine-tuning,
which significantly reduces the training cost of using generative large model
in wireless networks. Experimental results demonstrate that RadioDiff-Inverse
achieves state-of-the-art performance in accuracy of RM construction and
environmental reconstruction, and robustness against noisy sparse sampling.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:49:59 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Wang",
"Xiucheng",
""
],
[
"Fang",
"Zhongsheng",
""
],
[
"Cheng",
"Nan",
""
]
] |
2504.14299 | Matthew A. Kenworthy | M. A. Kenworthy, E. de Mooij, A. Brandeker, C. Opitom, F. Kiefer, A.
Fitzsimmons | Upper limits on CN from exocomets transiting $\beta$ Pictoris | 9 pages, 8 figures, 1 table, accepted for publication in A and A. The
paper is in a reproducible workflow repository at
https://github.com/mkenworthy/BetaPicCN | null | null | null | astro-ph.EP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The young (23 Myr) nearby (19.4 pc) star $\beta$ Pictoris hosts an edge-on
debris disk with two gas giant exoplanets in orbit around it. Many transient
absorption features have been detected in the rotationally broadened stellar
lines, which are thought to be the coma of infalling exocomets crossing the
line of sight towards Earth. In the Solar System, the molecule cynaogen (CN)
and its associated ionic species are one of the most detectable molecules in
the coma and tails of comets.
We perform a search for cyanogen in the spectra of $\beta$ Pictoris to detect
or put an upper limit on this molecule's presence in a young, highly active
planetary system. We divide twenty year's worth of HARPS spectra into those
with strong exocomet absorption features, and those with only stellar lines.
The high signal-to-noise stellar spectrum normalises out the stellar lines in
the exocomet spectra, which are then shifted and stacked on the deepest
exocomet absorption features to produce a high signal-to-noise exocomet
spectrum, and search for the CN band head using a model temperature dependent
cross-correlation template.
We do not detect CN in our data, and place a temperature and broadening
dependent 5$\sigma$ upper limit between 10$^{12}$ cm$^{-2}$ and 10$^{13}$
cm$^{-2}$, to be compared to the typical 10$^9$ - 10$^{10}$ cm$^{-2}$ expected
from scaling of the values in the Solar System comets.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 13:50:42 GMT"
}
] | 2025-04-22T00:00:00 | [
[
"Kenworthy",
"M. A.",
""
],
[
"de Mooij",
"E.",
""
],
[
"Brandeker",
"A.",
""
],
[
"Opitom",
"C.",
""
],
[
"Kiefer",
"F.",
""
],
[
"Fitzsimmons",
"A.",
""
]
] |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.