一作ACL 2026Workshop· Natural Language Processing
Processing Inconsistency Predicts Language Competence: LLM Evaluation Without Answer Labels on T urkic Languages
Ilya Galyukshev · Ilseyar Alimova
摘要
Most languages lack labeled evaluation benchmarks for large language models (LLMs). Creating such benchmarks requires native speakers, domain expertise, and answer annotation—resources unavailable for the vast majority of languages. We investigate whether a model’s internal processing signals—suc...