Papers
Research papers from arXiv and related sources
Invisible failures in human-AI interactions
AI systems fail silently far more often than they fail visibly. In a large-scale quantitative analysis of human-AI interactions from the WildChat dataset, we find that 78% of AI failures are invisi...
Christopher Potts, Moritz Sudhof
Amplification Effects in Test-Time Reinforcement Learning: Safety and Reasoning Vulnerabilities
Test-time training (TTT) has recently emerged as a promising method to improve the reasoning abilities of large language models (LLMs), in which the model directly learns from test data without acc...
Vanshaj Khattar, Md Rafi ur Rashid, Moumita Choudhury, Jing Liu, Toshiaki Koike-Akino, Ming Jin, ...
SEA-Vision: A Multilingual Benchmark for Comprehensive Document and Scene Text Understanding in Southeast Asia
Multilingual document and scene text understanding plays an important role in applications such as search, finance, and public services. However, most existing benchmarks focus on high-resource lan...
Pengfei Yue, Xingran Zhao, Juntao Chen, Peng Hou, Wang Longchao, Jianghang Lin, Shengchuan Zhang,...
TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems
With the rapid development of LLM-based multi-agent systems (MAS), their significant safety and security concerns have emerged, which introduce novel risks going beyond single agents or LLMs. Despi...
Kai Wang, Biaojie Zeng, Zeming Wei, Chang Jin, Hefeng Zhou, Xiangtian Li, Chao Yang, Jingjing Qu,...
Fusian: Multi-LoRA Fusion for Fine-Grained Continuous MBTI Personality Control in Large Language Models
Large Language Models (LLMs) have demonstrated impressive capabilities in simulating diverse human behaviors and personalities. However, existing methods for personality control, which include prom...
Zehao Chen, Rong Pan
A Closer Look into LLMs for Table Understanding
Despite the success of Large Language Models (LLMs) in table understanding, their internal mechanisms remain unclear. In this paper, we conduct an empirical study on 16 LLMs, covering general LLMs,...
Jia Wang, Chuanyu Qin, Mingyu Zheng, Qingyi Si, Peize Li, Zheng Lin
SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
Agent skills, structured procedural knowledge packages injected at inference time, are increasingly used to augment LLM agents on software engineering tasks. However, their real utility in end-to-e...
Tingxu Han, Yi Zhang, Wei Song, Chunrong Fang, Zhenyu Chen, Youcheng Sun, Lijie Hu
Spatial Characterization of Sub-Synchronous Oscillations Using Black-Box IBR Models
Power systems with high penetration of inverter-based resources (IBRs) are prone to sub-synchronous oscillations (SSO). The opaqueness of vendor-specific IBR models limits the ability to predict th...
Muhammad Sharjeel Javaid, Gabriel Covarrubias Maureira, Ambuj Gupta, Debraj Bhattacharjee, Jianli...
SFCoT: Safer Chain-of-Thought via Active Safety Evaluation and Calibration
Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks. However, they remain highly susceptible to jailbreak attacks that undermine their safety alignment...
Yu Pan, Wenlong Yu, Tiejun Wu, Xiaohu Ye, Qiannan Si, Guangquan Xu, Bin Wu
AI Evasion and Impersonation Attacks on Facial Re-Identification with Activation Map Explanations
Facial identification systems are increasingly deployed in surveillance and yet their vulnerability to adversarial evasion and impersonation attacks pose a critical risk. This paper introduces a no...
Noe Claudel, Weisi Guo, Yang Xing
When Does Sparsity Mitigate the Curse of Depth in LLMs
Recent work has demonstrated the curse of depth in large language models (LLMs), where later layers contribute less to learning and representation than earlier layers. Such under-utilization is lin...
Dilxat Muhtar, Xinyuan Song, Sebastian Pokutta, Max Zimmer, Nico Pelleriti, Thomas Hofmann, Shiwe...
RieMind: Geometry-Grounded Spatial Agent for Scene Understanding
Visual Language Models (VLMs) have increasingly become the main paradigm for understanding indoor scenes, but they still struggle with metric and spatial reasoning. Current approaches rely on end-t...
Fernando Ropero, Erkin Turkoz, Daniel Matos, Junqing Du, Antonio Ruiz, Yanfeng Zhang, Lu Liu, Min...
Why AI systems don't learn and what to do about it: Lessons on autonomous learning from cognitive science
We critically examine the limitations of current AI models in achieving autonomous learning and propose a learning architecture inspired by human and animal cognition. The proposed framework integr...
Emmanuel Dupoux, Yann LeCun, Jitendra Malik
More Test-Time Compute Can Hurt: Overestimation Bias in LLM Beam Search
Wider beam search should improve LLM reasoning, but when should you stop widening? Prior work on beam width selection has focused on inference efficiency \citep{qin2025dsbd, freitag2017beam}, witho...
Gal Dalal, Assaf Hallak, Gal Chechik, Yftach Ziser
Formalizing and validating properties in Asmeta with Large Language Models (Extended Abstract)
Writing temporal logic properties is often a challenging task for users of model-based development frameworks, particularly when translating informal requirements into formal specifications. In thi...
Andrea Bombarda, Silvia Bonfanti, Angelo Gargantini, Nico Pellegrinelli
GradCFA: A Hybrid Gradient-Based Counterfactual and Feature Attribution Explanation Algorithm for Local Interpretation of Neural Networks
Explainable Artificial Intelligence (XAI) is increasingly essential as AI systems are deployed in critical fields such as healthcare and finance, offering transparency into AI-driven decisions. Two...
Jacob Sanderson, Hua Mao, Wai Lok Woo
SKILLS: Structured Knowledge Injection for LLM-Driven Telecommunications Operations
As telecommunications operators accelerate adoption of AI-enabled automation, a practical question remains unresolved: can general-purpose large language model (LLM) agents reliably execute telecom...
Ivo Brett
Brain-Inspired Graph Multi-Agent Systems for LLM Reasoning
Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide range of language tasks, yet complex multi-step reasoning remains a fundamental challenge. While Large Reasoning...
Guangfu Hao, Yuming Dai, Xianzhe Qin, Shan Yu
CRASH: Cognitive Reasoning Agent for Safety Hazards in Autonomous Driving
As AVs grow in complexity and diversity, identifying the root causes of operational failures has become increasingly complex. The heterogeneity of system architectures across manufacturers, ranging...
Erick Silva, Rehana Yasmin, Ali Shoker
PMAx: An Agentic Framework for AI-Driven Process Mining
Process mining provides powerful insights into organizational workflows, but extracting these insights typically requires expertise in specialized query languages and data science tools. Large Lang...
Anton Antonov, Humam Kourani, Alessandro Berti, Gyunam Park, Wil M. P. van der Aalst