1 citations · 1 across the 15 of their papers we have counts for
15 papers
Taming CoT Obfuscation in VLMs: From Mechanistic Evidence to Activation-Level Enforcement
Xutao Mao, Jianing Zhu, Jinman Zhao +4
Reinforcement learning (RL) has substantially improved the reasoning capabilities of vision-language models (VLMs), yet it can induce chain-of-thought (CoT) obfuscation, an operati…
Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents
Xutao Mao, Liangjie Zhao, Xiang Zheng +1
LLM agents increasingly save experience from past tasks as reusable skills through skill evolution. When an agent completes an unsafe task, skill evolution can keep the unsafe proc…
Agents Don't Just Agree, They Remember: Benchmarking Persistent Sycophancy in Self-Improving Personal Agents
Xutao Mao, Liangjie Zhao, Leyao Wang +6
Self-improving personal agents write profiles, memories, and reusable skills that carry over from one chat to the next. Prior works ask whether a model agrees with the user in its…
Agent Hacks Agents: Autoresearch Discovers Vulnerabilities in Production Agents
Xutao Mao, Xiang Zheng, Rui Qian +1
Production LLM agents such as Claude Code and Codex can modify files and execute commands, so safety failures become real destructive actions. Automatic red-teaming lets safety tea…
Sandlock: Confining AI Agent Code with Unprivileged Linux Primitives
Cong Wang, Yusheng Zheng
AI agents increasingly run untrusted code on developer machines: shell commands generated by language models, third-party scripts retrieved at runtime, and tool plugins of unknown…
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
Xiao Li, Xiang Zheng, Yifeng Gao +35
Embodied Artificial Intelligence (Embodied AI) integrates perception, cognition, planning, and interaction into agents that operate in open-world, safety-critical environments. As…
What Happens Inside Agent Memory? Circuit Analysis from Emergence to Diagnosis
Xutao Mao, Jinman Zhao, Gerald Penn +1
Agent memory failures are silent: an LLM-based agent can produce a fluent response even when it fails to extract, retain, or retrieve the information needed across sessions. The wr…
HazardArena: Evaluating Semantic Safety in Vision-Language-Action Models
Zixing Chen, Yifeng Gao, Li Wang +8
Vision-Language-Action (VLA) models inherit rich world knowledge from vision-language backbones and acquire executable skills from action demonstrations. Yet current evaluations pr…
Internal Safety Collapse in Frontier Large Language Models
Yutao Wu, Xiao Liu, Yifeng Gao +7
This work identifies a critical failure mode in frontier large language models (LLMs), which we term Internal Safety Collapse (ISC): under certain task conditions, models enter a s…
Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs
Xiang Zheng, Yutao Wu, Hanxun Huang +5
Autonomous code agents built on large language models are reshaping software and AI development through tool use, long-horizon reasoning, and self-directed interaction. However, th…
RedTopic: Toward Topic-Diverse Red Teaming of Large Language Models
Jiale Ding, Xiang Zheng, Yutao Wu +5
As large language models (LLMs) are increasingly deployed as black-box components in real-world applications, red teaming has become essential for identifying potential risks. It t…
RedRFT: A Light-Weight Benchmark for Reinforcement Fine-Tuning-Based Red Teaming
Xiang Zheng, Xingjun Ma, Wei-Bin Lee +1
Red teaming has proven to be an effective method for identifying and mitigating vulnerabilities in Large Language Models (LLMs). Reinforcement Fine-Tuning (RFT) has emerged as a pr…
Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
Xingjun Ma, Yifeng Gao, Yixu Wang +45
The rapid advancement of large models, driven by their exceptional abilities in learning and generalization through large-scale pre-training, has reshaped the landscape of Artifici…
CALM: Curiosity-Driven Auditing for Large Language Models
Xiang Zheng, Longxiang Wang, Yi Liu +3
Auditing Large Language Models (LLMs) is a crucial and challenging task. In this study, we focus on auditing black-box LLMs without access to their parameters, only to the provided…
Constrained Intrinsic Motivation for Reinforcement Learning
Xiang Zheng, Xingjun Ma, Chao Shen +1
This paper investigates two fundamental problems that arise when utilizing Intrinsic Motivation (IM) for reinforcement learning in Reward-Free Pre-Training (RFPT) tasks and Explora…