activity
20242026
most citedSaRO: Enhancing LLM Safety through Reasoning-based Alignment

1 citations · 3 across the 13 of their papers we have counts for

collaborators

19 papers

cs.LG2026

What Do Agents Learn from Trajectory-SFT: Semantics or Interfaces?

Weizheng Gu, Chengze Li, Zhuohao Yu +6

Large language models are increasingly evaluated as interactive agents, yet standard agent benchmarks conflate two qualitatively distinct sources of success: semantic tool-use and…

cs.CL2026

ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback

Yutao Mou, Zhangchi Xue, Lijun Li +4

While LLM-based agents can interact with environments via invoking external tools, their expanded capabilities also amplify security risks. Monitoring step-level tool invocation be…

cs.CL2025

Modeling Uncertainty Trends for Timely Retrieval in Dynamic RAG

Bo Li, Tian Tian, Zhenghua Xu +3

Dynamic retrieval-augmented generation (RAG) allows large language models (LLMs) to fetch external knowledge on demand, offering greater adaptability than static RAG. A central cha…

cs.SE2025

Benchmarking and Studying the LLM-based Agent System in End-to-End Software Development

Zhengran Zeng, Yixin Li, Rui Xie +2

The development of LLM-based autonomous agents for end-to-end software development represents a significant paradigm shift in software engineering. However, the scientific evaluati…

cs.CL2025

Decoupling Safety into Orthogonal Subspace: Cost-Efficient and Performance-Preserving Alignment for Large Language Models

Yutao Mou, Xiaoling Zhou, Yuxiao Luo +2

Safety alignment is essential for building trustworthy artificial intelligence, yet it remains challenging to enhance model safety without degrading general performance. Current ap…

cs.CL2025

AutoRed: A Free-form Adversarial Prompt Generation Framework for Automated Red Teaming

Muxi Diao, Yutao Mou, Keqing He +6

The safety of Large Language Models (LLMs) is crucial for the development of trustworthy AI applications. Existing red teaming methods often rely on seed instructions, which limits…