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

Xiaozhi Wang

2 indexed papers

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

Publications per year

2
26

Top categories

AI×2ML×1NLP×1

Frequent co-authors

Hao Peng2×
Juanzi Li2×
Yunjia Qi1×
Zehua Yin1×
Xintong Shi1×
Songyuanyi Lu1×

Research Timeline

2026
Reproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement Learning

This paper introduces CHERRL, a controllable hacking environment for rubric-based reinforcement learning to study and mitigate reward hacking.

TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories

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.

Highlighted terms show continued research focus across papers

Papers

cs.AIEmpiricalRecentAug 6, 2026

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…

View →
cs.LGcs.AIcs.CLRecent
Jun 3, 2026

Reproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement Learning

Xuekang Wang, Zhuoyuan Hao, Shuo Hou, Hao Peng +2 more

This paper introduces CHERRL, a controllable hacking environment for rubric-based reinforcement learning to study and mitigate reward hacking.

View →