conftrace_

Wangyan

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

Conferences

ACL (2)

Top co-authors

Zhang Yueyang (2) Zhiyuan Sun (2) Daiting Shi (2) Liangxin Liu (2) Longzheng Wang (2) Yu Liang (2) Long Xia (2) Houde Liu (1) Kai Qin (1)

Keywords

generative reward model (2) reinforcement learning from human feedback (2) preference alignment (1) large language model (1) consistency-aware reward (1) positional bia (1) preference modeling (1)

Papers

ReflectRM: Boosting Generative Reward Models via Self-Reflection within a Unified Judgment Framework ACL 2026 ConsistRM: Improving Generative Reward Models via Consistency-Aware Self-Training ACL 2026