Skip to content
NOTES
MenuClose
NOTES // concept

LLM-as-a-judge

LAST_MODIFIED
2026.07.20
CATEGORY
concept

LLM-as-a-judge - a model grades another model’s output (or full trajectory) against a written rubric, instead of exact-match metrics like BLEU/ROUGE that ignore semantics.

Which judge shape

NeedPrefer
Clear success conditionDAG (deterministic yes/no tree)
Subjective qualityG-Eval (steps then score)
Fact-check against sourcesQAG (closed yes/no questions)

Self-evaluation (same model scoring its own answer) is calibration, not introspection - Ren et al. showed it beats sequence-likelihood baselines on TruthfulQA and TL;DR.

On this site

The Claude Code /goal path is production LLM-as-a-judge with a hard gate: the evaluator only reads the transcript and a verifiable completion condition (transcript-only rule). Full pipeline sizing: Five metrics for an agent eval pipeline.

evalsllm-as-judgeagentsquality

Related_Notes