Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Zhaojiacheng Zhou

Zhaojiacheng Zhou

2 indexed papers

Recent (6 mo)
2
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

Crypto×2AI×2ML×1

Frequent co-authors

Kaixiang Wang1×
Jiong Lou1×
Jie Li1×

Research Timeline

2026
Proteus: A Self-Evolving Red Team for Agent Skill Ecosystems

The paper introduces Proteus, a self-evolving red-team framework that measures the adaptive leakage risk of LLM agent skills, demonstrating that current vetting methods significantly underestimate residual risk against iterative attackers.

OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences

The paper introduces Obsessive Experience Poisoning (OEP), a low-privilege black-box attack that poisons self-evolving LLM agents by generating locally correct but harmful experiences, causing dangerous over-generalization during reflection.

Highlighted terms show continued research focus across papers

Papers

cs.CRcs.AIcs.LGRecentMay 18, 2026

OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences

Kaixiang Wang, Jiong Lou, Zhaojiacheng Zhou, Jie Li

The paper introduces Obsessive Experience Poisoning (OEP), a low-privilege black-box attack that poisons self-evolving LLM agents by generating locally correct but harmful experiences, causing dangero…

View →
cs.CRcs.AIRecentMay 12, 2026

Proteus: A Self-Evolving Red Team for Agent Skill Ecosystems

Zhaojiacheng Zhou

The paper introduces Proteus, a self-evolving red-team framework that measures the adaptive leakage risk of LLM agent skills, demonstrating that current vetting methods significantly underestimate res…

View →