作者位次 #5/6ACL 2026Poster· Natural Language Processing
Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM -as-a-Judge
Xin Sun · Di Wu · Sijing Qin · Isao Echizen · Abdallah El Ali · Saku Sugawara
摘要
Large language models (LLMs) are increasingly used as automated evaluators (LLM-as-a-Judge). This work challenges its reliability by showing that trust judgments by LLMs are biased by disclosed source labels. Using a counterfactual design, we find that both humans and LLM judges assign higher tru...