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

Pan Zhou

2 indexed papers

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

Publications per year

2
26

Top categories

ML×1NLP×1Crypto×1AI×1

Frequent co-authors

Minghui Zheng1×
Hongxu Chen1×
Huimin Ren1×
Hongsheng Xin1×
Xiaoyang Qu1×
Ze Wang1×

Research Timeline

2026
BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning

The paper introduces BadSkill, a novel backdoor attack formulation that targets third-party agent skills by poisoning the embedded model artifacts, achieving high attack success rates across various model sizes.

HMPO: Hybrid Median-length Policy Optimization for Chain-of-Thought Compression

HMPO introduces a single-stage, cost-effective reinforcement learning framework that achieves significant token compression of Chain-of-Thought reasoning with minimal loss of accuracy, applicable across various large language model architectures.

Highlighted terms show continued research focus across papers

Papers

cs.LGcs.CLRecentJun 1, 2026

HMPO: Hybrid Median-length Policy Optimization for Chain-of-Thought Compression

Minghui Zheng, Hongxu Chen, Huimin Ren, Hongsheng Xin +7 more

HMPO introduces a single-stage, cost-effective reinforcement learning framework that achieves significant token compression of Chain-of-Thought reasoning with minimal loss of accuracy, applicable acro…

View →
cs.CRcs.AIRecentApr 10, 2026

BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning

Guiyao Tie, Jiawen Shi, Pan Zhou, Lichao Sun

The paper introduces BadSkill, a novel backdoor attack formulation that targets third-party agent skills by poisoning the embedded model artifacts, achieving high attack success rates across various m…

View →