English
Related papers

Related papers: Policy-Invisible Violations in LLM-Based Agents

200 papers

The rise of large language model (LLM)-based multi-agent systems (MAS) introduces new security and reliability challenges. While these systems show great promise in decomposing and coordinating complex tasks, they also face multi-faceted…

Artificial Intelligence · Computer Science 2025-06-02 Xu He , Di Wu , Yan Zhai , Kun Sun

Multimodal large language models (MLLMs) are increasingly deployed in open-ended, real-world environments where inputs are messy, underspecified, and not always trustworthy. Unlike curated benchmarks, these settings frequently involve…

Artificial Intelligence · Computer Science 2025-08-26 Qianqi Yan , Hongquan Li , Shan Jiang , Yang Zhao , Xinze Guan , Ching-Chen Kuo , Xin Eric Wang

The rapid adoption of LLM-based agentic systems has produced a rich ecosystem of frameworks (smolagents, LangGraph, AutoGen, CAMEL, LlamaIndex, i.a.). Yet existing benchmarks are model-centric: they fix the agentic setup and do not compare…

Artificial Intelligence · Computer Science 2026-03-11 Cornelius Emde , Alexander Rubinstein , Anmol Goel , Ahmed Heakl , Sangdoo Yun , Seong Joon Oh , Martin Gubri

Large Language Model (LLM) agents offer a powerful new paradigm for solving various problems by combining natural language reasoning with the execution of external tools. However, their dynamic and non-transparent behavior introduces…

Cryptography and Security · Computer Science 2025-11-19 Peiran Wang , Yang Liu , Yunfei Lu , Yifeng Cai , Hongbo Chen , Qingyou Yang , Jie Zhang , Jue Hong , Ye Wu

We introduce PentestJudge, a system for evaluating the operations of penetration testing agents. PentestJudge is a large language model (LLM)-as-judge with access to tools that allow it to consume arbitrary trajectories of agent states and…

Artificial Intelligence · Computer Science 2025-08-06 Shane Caldwell , Max Harley , Michael Kouremetis , Vincent Abruzzo , Will Pearce

As multimodal large language models (MLLMs) advance, MLLM-based virtual agents have demonstrated remarkable performance. However, existing benchmarks face significant limitations, including uncontrollable task complexity, extensive manual…

Computer Vision and Pattern Recognition · Computer Science 2025-06-11 Wendong Bu , Yang Wu , Qifan Yu , Minghe Gao , Bingchen Miao , Zhenkui Zhang , Kaihang Pan , Yunfei Li , Mengze Li , Wei Ji , Juncheng Li , Siliang Tang , Yueting Zhuang

Agents built on LLMs are increasingly deployed across diverse domains, automating complex decision-making and task execution. However, their autonomy introduces safety risks, including security vulnerabilities, legal violations, and…

Artificial Intelligence · Computer Science 2025-08-01 Haoyu Wang , Christopher M. Poskitt , Jun Sun

Reinforcement Learning (RL) agents deployed in real-world environments face degradation from sensor faults, actuator wear, and environmental shifts, yet lack intrinsic mechanisms to detect and diagnose these failures. We present an…

Artificial Intelligence · Computer Science 2025-09-15 Cameron Reid , Wael Hafez , Amirhossein Nazeri

Existing LLM-based agent systems share a common architectural failure: they answer from the unrestricted knowledge space without first simulating how active business scenarios reshape that space for the event at hand -- producing decisions…

Artificial Intelligence · Computer Science 2026-04-13 Hongyin Zhu , Jinming Liang , Mengjun Hou , Ruifan Tang , Xianbin Zhu , Jingyuan Yang , Yuanman Mao , Feng Wu

People increasingly use LLM agents for multi-turn financial recommendations, where the agent pulls market data through tools and tracks user preferences across turns. When tool outputs are manipulated, the recommendations stop matching the…

Computation and Language · Computer Science 2026-05-27 Zekun Wu , Adriano Koshiyama , Sahan Bulathwela , Maria Perez-Ortiz

Multi-agent systems built on large language models (LLMs) are expected to enhance decision-making by pooling distributed information, yet systematically evaluating this capability has remained challenging. We introduce HiddenBench, a…

Computation and Language · Computer Science 2026-05-14 Yuxuan Li , Aoi Naito , Hirokazu Shirado

High-risk industries like nuclear and aviation use real-time monitoring to detect dangerous system conditions. Similarly, Large Language Models (LLMs) need monitoring safeguards. We propose a real-time framework to predict harmful AI…

Artificial Intelligence · Computer Science 2025-05-21 Maheep Chaudhary , Fazl Barez

We present a modular, explainable LLM-agent pipeline for decision support that externalizes reasoning into auditable artifacts. The system instantiates three frameworks: Vester's Sensitivity Model (factor set, signed impact matrix, systemic…

Artificial Intelligence · Computer Science 2025-11-11 Marcel Pehlke , Marc Jansen

Behavioral evaluation is the dominant paradigm for assessing alignment in large language models (LLMs). In current practice, observed compliance under finite evaluation protocols is treated as evidence of latent alignment. However, the…

Machine Learning · Computer Science 2026-02-10 Igor Santos-Grueiro

Existing research primarily evaluates the values of LLMs by examining their stated inclinations towards specific values. However, the "Value-Action Gap," a phenomenon rooted in environmental and social psychology, reveals discrepancies…

Human-Computer Interaction · Computer Science 2025-10-01 Hua Shen , Nicholas Clark , Tanushree Mitra

Multi-agent systems achieve state-of-the-art outcomes through peer collaboration. However, when an agent in the pipeline silently drops a constraint, the system's final output may look correct even though the reasoning chain was quietly…

Monitoring autonomous large language model (LLM) agents for covert malicious behavior is challenging due to delayed, context-dependent, and long-horizon attack patterns. Agents may pursue hidden objectives while maintaining superficially…

Machine Learning · Computer Science 2026-05-26 Nesreen K. Ahmed , Nima Nafisi

The widespread use of cameras in our society has created an overwhelming amount of video data, far exceeding the capacity for human monitoring. This presents a critical challenge for public safety and security, as the timely detection of…

Computer Vision and Pattern Recognition · Computer Science 2025-10-28 Pascal Benschop , Cristian Meo , Justin Dauwels , Jelte P. Mense

Defenses against indirect prompt injection (IPI) in tool-using LLM agents share two structural weaknesses. First, they all attempt to prevent attacks rather than detect the compromises that slip through. Second, they have only been…

Cryptography and Security · Computer Science 2026-05-13 Yassin H. Rassul , Tarik A. Rashid

Financial agents powered by large language models (LLMs) are increasingly deployed for investment analysis, risk assessment, and automated decision-making, where their abilities to plan, invoke tools, and manipulate mutable state introduce…