← Search

Yishan Li

2 accepted papers

2026

SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems

ICML 2026poster

Skill-based agent systems tackle complex tasks by composing reusable skills, improving modularity and scalability while introducing a largely unexamined security attack surface. We propose \textbf{SkillTrojan}, a backdoor attack that targets skill implementations rather than model parameters or trai…

Cited by 0SourceScholar
2025

RAGEval: Scenario Specific RAG Evaluation Dataset Generation Framework

ACL 2025long

Retrieval-Augmented Generation (RAG) is a powerful approach that enables large language models (LLMs) to incorporate external knowledge. However, evaluating the effectiveness of RAG systems in specialized scenarios remains challenging due to the high costs of data construction and the lack of suitab…