conftrace_

Qipeng Huang

2 papers · 2026–2026 · 1 conference · across top CS/AI conferences

Conferences

ACL (2)

Top co-authors

Min Zhang (2) Yuyang Ding (2) xiaobo liang (2) Juntao Li (2) Wanfu Wang (1) Zhe Zhao (1) Zecheng Tang (1) Wenpeng Zhu (1) Qianben Chen (1) Zhang Yijun (1)

Keywords

reinforcement learning (1) preference alignment (1) cross-modal transfer (1) credit assignment (1) test-time scaling (1) generative reward model (1) exploration collapse (1) behavior collapse (1)

Papers

Escaping the Echo Trap: On Credit Assignment Failure in Multi-turn LLM Self-Reflection ACL 2026 DUAL RM: Beyond Rule-based Preference Reward Modeling via Meta-Reward ACL 2026