作者位次 #2/8ACL 2026OralBest Resource Paper· Natural Language Processing
HSC ode C omp: A Realistic and Expert-level Agent Benchmark for Hierarchical Rule Application
Tian Lan · Yiqian Yang · Qianghuai Jia · Li Zhu · Hui Jiang · Hang Zhu · Weihua Luo · Longyue Wang
摘要
Despite recent progress, existing agent benchmarks neglect a fundamental real-world capability: hierarchical rule application, a critical requirement in fields such as law and medicine where agents must reason from broad categories down to specific exceptions to reach rule-compliant decisions.Thi...