EMNLP Paper Questions LLM-as-a-Judge Reliability Due to Rubric Artifacts
A paper accepted at EMNLP 2026 shows that classifiers trained only on rubric text, without seeing the evaluated answers, can partially predict LLM judge scores, revealing exploitable rubric artifacts that undermine LLM-as-a-Judge reliability.
2026-10-27 ~ 2026-10-28 · 2 related posts
- EMNLP Poster: LLM-as-a-Judge Verdicts Track Rubric Artifacts, Not Real Quality — ravi_iitm · 2026-10-27
- EMNLP paper: rubric text alone can predict LLM-as-a-Judge scores — ravi_iitm · 2026-10-28