作者位次 #2/7ACL 2026Findings· Findings of ACL
Breaking the Evaluation Paradox: Evaluating High-Entropy Search with Computationally Irreducible Constraints
Juntao Wu · Wei Wen · Xianting Huang · Shuai Pang · Ruizhi Qiao · Xing Sun · Ke Wang
摘要
Evaluating the exhaustive search capabilities of large language models (LLMs) is plagued by a fundamental paradox: verifying completeness requires complete ground truth, yet high-entropy enumeration tasks make such ground truth impossible for humans to create. This causes benchmarks to systematic...