作者位次 #7/11ACL 2026Findings· Findings of ACL
AJ -Bench: Benchmarking Agent-as-a-Judge for Environment-Aware Evaluation
Wentao Shi · Yu Wang · Yuyang Zhao · Yuxin Chen · Fuli Feng · Xueyuan Hao · Xi Su · Qi GU · Hui Su · Xunliang Cai · Xiangnan He
摘要
As reinforcement learning continues to scale the training of large language model–based agents, reliably verifying agent behaviors in complex environments has become increasingly challenging. Existing approaches rely on rule-based verifiers or LLM-as-a-Judge models, which struggle to generalize b...