What Is LLM-as-a-Judge? and How It Works
LLM-as-a-judge is an evaluation technique where one large language model assesses the output of another based on specific natural language criteria. This method has become a standard for scaling quality checks in production environments because it approximates human judgment more effectively than traditional text metrics. However, the approach requires careful calibration to mitigate inherent biases such as length preference and positional bias to ensure reliable results.
- ▪LLM-as-a-judge allows teams to evaluate open-ended model outputs at production volume without the need for retraining.
- ▪The method supports development comparisons, live monitoring, and regression checks by using prompts to define evaluation criteria.
- ▪Judges can exhibit measurable biases, including rewarding longer responses and favoring the first option in pairwise comparisons.
- ▪Proper implementation requires a well-built evaluation prompt, a measured baseline, and a calibration step against human labels.
Hacker News (AI / LLM) files mainly under ai. We currently carry 6,402 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Infere |
| Canonical URL | https://infere.com/blog/what-is-llm-as-a-judge/ |
| Publication time | Fri, 25 Sep 2026 07:19:46 +0000 |
| Retrieval time | 2026-09-25T07:20:28.329Z |
| Last seen | 2026-09-25T07:20:28.329Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | jLhK8Wf8cHfr · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
LLM-as-a-judge is an evaluation method where one large language model scores or compares the text that another LLM produces. You give the judge model a short prompt that states the criteria you care about, the question, and the answer to grade, and the judge returns a number, a label, or a preference between two answers, usually with a written reason for its verdict. It exists because human review does not scale and because classic text metrics miss meaning, and it has become the default way teams evaluate open-ended model output, chatbot conversations, and agent behavior at production volume. The honest part comes next. An LLM judge is not a metric you switch on and trust. It is an automated stand-in for a human annotator, and it carries real, measurable biases.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Infere.
Discussion
0 commentsMore from Infere
