一作ACL 2026Findings· Findings of ACL
No-Worse Context-Aware Decoding: Preventing Neutral Regression in Context-Conditioned Generation
Yufei Tao · Ameeta Agrawal
摘要
Large language models (LLMs) can answer questions and summarize documents when conditioned on external contexts (e.g., retrieved evidence), yet context use remains unreliable: models may overwrite an already-correct output (neutral regression) even when the context is non-informative. We formaliz...