作者位次 #9/16ACL 2026Findings· Findings of ACL
A gent V - RL : Scaling Reward Modeling with Agentic Verifier
Jiazheng Zhang · Ziche Fu · Zhiheng Xi · Wenqing Jing · Mingxu Chai · Wei He · Guoqiang Zhang · Chenghao Fan · Chenxin An · Wenxiang Chen · Zhicheng Liu · Haojie Pan · Dingwei Zhu · Tao Gui · Qi Zhang · Xuanjing Huang
摘要
Verifiers have been demonstrated to enhance LLM reasoning via test-time scaling (TTS). Yet, they face significant challenges in complex domains. Error propagation from incorrect intermediate reasoning can lead to false positives for seemingly plausible solutions, while lacking external grounding...