Research

Papers

Research papers from arXiv and related sources

Total: 4513 AI/LLM: 2483 Testing: 2030
TESTING

HEP Statistical Inference for UAV Fault Detection: CLs, LRT, and SBI Applied to Blade Damage

This paper transfers three statistical methods from particle physics to multirotor propeller fault detection: the likelihood ratio test (LRT) for binary detection, the CLs modified frequentist meth...

Khushiyant

2603.18546 2026-03-19
TESTING

CoDA: Exploring Chain-of-Distribution Attacks and Post-Hoc Token-Space Repair for Medical Vision-Language Models

Medical vision--language models (MVLMs) are increasingly used as perceptual backbones in radiology pipelines and as the visual front end of multimodal assistants, yet their reliability under real c...

Xiang Chen, Fangfang Yang, Chunlei Meng, Chengyin Hu, Ang Li, Yiwei Wei, Jiahuan Long, Jiujiang Guo

2603.18545 2026-03-19
TESTING

Beyond Passive Aggregation: Active Auditing and Topology-Aware Defense in Decentralized Federated Learning

Decentralized Federated Learning (DFL) remains highly vulnerable to adaptive backdoor attacks designed to bypass traditional passive defense metrics. To address this limitation, we shift the defens...

Sheng Pan, Niansheng Tang

2603.18538 2026-03-19
TESTING

When Names Change Verdicts: Intervention Consistency Reveals Systematic Bias in LLM Decision-Making

Large language models (LLMs) are increasingly used for high-stakes decisions, yet their susceptibility to spurious features remains poorly characterized. We introduce ICE-Guard, a framework applyin...

Abhinaba Basu, Pavan Chakraborty

2603.18530 2026-03-19
TESTING

Wind accretion onto planets orbiting an evolving Solar-like star and their detectability

As stars evolve, they undergo significant changes in their physical properties, which can have a profound impact on the planets orbiting them. In particular, the mass lost through stellar wind may ...

P. Padilla-López, R. F. Maldonado, J. A. Toalá, E. Tejeda, J. B. Rodríquez-González

2603.18525 2026-03-19
TESTING

Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models

Counting serves as a simple but powerful test of a Large Vision-Language Model's (LVLM's) reasoning; it forces the model to identify each individual object and then add them all up. In this study, ...

Liwei Che, Zhiyu Xue, Yihao Quan, Benlin Liu, Zeru Shi, Michelle Hurst, Jacob Feldman, Ruixiang T...

2603.18523 2026-03-19
TESTING

A systematic search for physical associations between fast radio bursts and astrophysical transients

The physical origin of fast radio bursts (FRBs) remains an unsolved mystery in astrophysics, with the magnetar central engine model as the leading framework. Systematically searching for physical a...

Hao-Hao Chen, Wen-Tao Xu, Xin-Yu Liang, Ming-Xuan Lu, Can-Min Deng

2603.18487 2026-03-19
TESTING

Leveraging Large Language Models for Generalizing Peephole Optimizations

Peephole optimizations are a core component of modern optimizing compilers. It rewrites specific instruction into semantically equivalent but more efficient forms. In practice, creating a new peeph...

Chunhao Liao, Hongxu Xu, Xintong Zhou, Zhenyang Xu, Chengnian Sun

2603.18477 2026-03-19
TESTING

Modeling Adversarial Wildfires for Power Grid Disruption

Electric power infrastructure faces increasing risk of damage and disruption due to wildfire. Operators of power grids in wildfire-prone regions must consider the potential impacts of unpredictable...

Matthew Brun, Xu Andy Sun, Jean-Paul Watson

2603.18473 2026-03-19
TESTING

$d_{N Ω}$ production in $Ωd$ scattering process

In the present work, we propose to investigate the production of $d_{N Ω}$ in the $Ω^{-} d \rightarrow p d_{N Ω}^-$ process by utilizing an effective Lagrangian approach, where $d_{N Ω}$ is identif...

Quan-Yun Guo, Jing Liu, Dian-Yong Chen

2603.18467 2026-03-19
TESTING

Learning Consistent Temporal Grounding between Related Tasks in Sports Coaching

Video-LLMs often attend to irrelevant frames, which is especially detrimental for sports coaching tasks requiring precise temporal grounding. Yet obtaining frame-level supervision is challenging: e...

Arushi Rai, Adriana Kovashka

2603.18453 2026-03-19
TESTING

AS2 -- Attention-Based Soft Answer Sets: An End-to-End Differentiable Neuro-Soft-Symbolic Reasoning Architecture

Neuro-symbolic artificial intelligence (AI) systems typically couple a neural perception module to a discrete symbolic solver through a non-differentiable boundary, preventing constraint-satisfacti...

Wael AbdAlmageed

2603.18436 2026-03-19
TESTING

Adaptive Decoding via Test-Time Policy Learning for Self-Improving Generation

Decoding strategies largely determine the quality of Large Language Model (LLM) outputs, yet widely used heuristics such as greedy or fixed temperature/top-p decoding are static and often task-agno...

Asmita Bhardwaj, Yuya Jeremy Ong, Eelaaf Zahid, Basel Shbita

2603.18428 2026-03-19
TESTING

The Impact of Corporate AI Washing on Farmers' Digital Financial Behavior Response -- An Analysis from the Perspective of Digital Financial Exclusion

In the context of the rapid development of digital finance, some financial technology companies exhibit the phenomenon of "AI washing," where they overstate their AI capabilities while underinvesti...

Li Wenxiu, Wen Zhanjie, Xia Jiechang, Guo Jingqiao

2603.18421 2026-03-19
TESTING

The Spillover Effects of Peer AI Rinsing on Corporate Green Innovation

At a time when the phenomenon of 'AI washing' is quietly spreading, an increasing number of enterprises are using the label of artificial intelligence merely as a cosmetic embellishment in their an...

Li Wenxiu, Wen Zhanjie, Xia Jiechang, Guo Jingqiao

2603.18415 2026-03-19
TESTING

Statistical Testing Framework for Clustering Pipelines by Selective Inference

A data analysis pipeline is a structured sequence of steps that transforms raw data into meaningful insights by integrating multiple analysis algorithms.In many practical applications, analytical f...

Yugo Miyata, Tomohiro Shiraishi, Shunichi Nishino, Ichiro Takeuchi

2603.18413 2026-03-19
TESTING

TARo: Token-level Adaptive Routing for LLM Test-time Alignment

Large language models (LLMs) exhibit strong reasoning capabilities but typically require expensive post-training to reach high performance. Recent test-time alignment methods offer a lightweight al...

Arushi Rai, Qiang Zhang, Hanqing Zeng, Yunkai Zhang, Dipesh Tamboli, Xiangjun Fan, Zhuokai Zhao

2603.18411 2026-03-19
TESTING

Where are the Hidden Gems? Applying Transformer Models for Design Discussion Detection

Design decisions are at the core of software engineering and appear in Q\&A forums, mailing lists, pull requests, issue trackers, and commit messages. Design discussions spanning a project's histor...

Lawrence Arkoh, Daniel Feitosa, Wesley K. G. Assunção

2603.18393 2026-03-19
TESTING

Reflection in the Dark: Exposing and Escaping the Black Box in Reflective Prompt Optimization

Automatic prompt optimization (APO) has emerged as a powerful paradigm for improving LLM performance without manual prompt engineering. Reflective APO methods such as GEPA iteratively refine prompt...

Shiyan Liu, Qifeng Xia, Qiyun Xia, Yisheng Liu, Xinyu Yu, Rui Qu

2603.18388 2026-03-19
TESTING

A Non-parametric Method for the Inference of Halo Occupation Distributions

The galaxy-halo connection traces processes by which galaxies form and evolve. The halo occupation distribution (HOD) describes the relationship between galaxies and their host dark matter haloes. ...

Jacob Kennedy, Eric Gawiser, Kartheik G. Iyer, L. Y. Aaron Yung

2603.18379 2026-03-19