From the 2 of 8 linked papers with an AI index.
8 papers
Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents
Xutao Mao, Liangjie Zhao, Xiang Zheng +1
Self-improving LLM agents convert successful trajectories into persistent cross-task state. An unsafe success can thereby become reusable policy after its triggering input disappea…
Agents Don't Just Agree, They Remember: Benchmarking Persistent Sycophancy in Stateful Personal Agents
Xutao Mao, Liangjie Zhao, Leyao Wang +6
The paper defines persistent sycophancy, where personal agents store user‑provided claims in long‑term memory and later repeat them, and introduces the Personal Agent Sycophancy Be…
Agent Hacks Agent: Autoresearch for Production-Agent Red-Teaming
Xutao Mao, Xiang Zheng, Cong Wang
The paper introduces AHA, an automated system that discovers and documents reusable vulnerability concepts in production LLM agents by hypothesizing, testing, and recording unsafe…
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…
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…
STARE: Step-wise Temporal Alignment and Red-teaming Engine for Multi-modal Toxicity Attack
Xutao Mao, Liangjie Zhao, Tao Liu +3
Red-teaming Vision-Language Models is essential for identifying vulnerabilities where adversarial image-text inputs trigger toxic outputs. Existing approaches treat image generatio…