Xiaozhi Wang
2 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
This paper introduces CHERRL, a controllable hacking environment for rubric-based reinforcement learning to study and mitigate reward hacking.
This paper introduces TrajDebug, an error-lifecycle tracing framework for LLM-based agentic systems, which addresses long-trajectory error discovery using multi-granularity history compression and evidence-based error identification.
Papers
TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
Yunjia Qi, Zehua Yin, Xintong Shi, Hao Peng +9 more
This paper introduces TrajDebug, an error-lifecycle tracing framework for LLM-based agentic systems, which addresses long-trajectory error discovery using multi-granularity history compression and evi…