liveBook · Manning

SelfCheckGPT

As explained in LLM Evaluation and Alignment, The Foundational Ideas

SelfCheckGPT is a hallucination-detection method that estimates factual reliability from the consistency of independently sampled model statistics or outputs.

A fluent answer can contain invented facts even when it sounds plausible. SelfCheckGPT estimates hallucination risk by comparing an answer with independently sampled responses to the same prompt. It needs no external knowledge base or access to model internals, making it useful for open-domain, black-box models. Agreement is a signal of reliability, not proof of truth. section 5.0 section 5.2 section 5.2.5

To check a response:

  • Generate the target response to the prompt. section 5.2.2
  • Generate N additional responses independently from the same prompt, using a temperature above zero. Do not expose any sample to the target or other samples. section 5.2.2
  • Compare each target sentence with the samples. Consistent sentences receive lower hallucination-risk scores; contradictions or missing support can raise the score, depending on the comparison method. section 5.2.2 section 5.2.3

Start with 5–10 samples for a small development test or 10–20 for offline evaluation. More samples can expose subtler variation, but each adds generation cost and latency. The original evaluation used 20 samples. section 5.2.2 section 5.2.4

One implementation uses BERTScore to compare each target sentence with every sentence in a sample. It takes the best match from each sample, averages those matches, and subtracts the average from 1. Higher scores indicate greater estimated risk. BERTScore measures semantic similarity, however, and can miss contradictions between similar statements. section 5.2.3

Scoring a sentence

If the best matches for a target sentence score 0.90, 0.80, and 0.40 across three samples, their average is 0.70. The inverted hallucination-risk score is 1 − 0.70 = 0.30. This is a screening signal, not a verified probability that the sentence is false. section 5.2.3

Table 5.2 Comparison of SelfCheckGPT variants
Variant
Accuracy
Cost
Latency
Best for
BERTScore
Moderate
Low
Fast
High-volume screening
NLI
Good
Low-Medium
Medium
Detecting contradictions
QA
Good
Medium
Medium
Fine-grained claim verifications
LLM-as-a-Judge
Highest
High
Slow
High-stakes evaluation

Compare BERTScore, natural language inference, question answering, and LLM-as-a-judge variants by their reported cost, latency, and use cases. section 5.2.3

The method can miss a false belief repeated consistently across samples. It can also flag a whole sentence without identifying which embedded claim is unsupported. For auditable verification, decompose responses into claims and check them against evidence when trusted evidence is available. section 5.2.5 section 5.1.4

In an evaluation on WikiBio, the LLM-as-a-judge variant achieved 0.832 sentence-level AUC-PR, a measure of precision and recall across detection thresholds. This result is specific to that evaluation, not a guarantee for every deployment. section 5.2.5

How LLM Evaluation and Alignment, The Foundational Ideas covers it

How deep it goes
Explains independent sampling, sentence-level scoring, implementation variants, sample-count trade-offs, and detection limitations.
Examples
Uses a Pantheon response, a three-sample BERTScore calculation, and a WikiBio evaluation to illustrate checking and results.
What you can do after
Evaluate/understand sentence-level hallucination risk by sampling responses, comparing consistency, and interpreting scores alongside their limitations.
How it connects
Connects hallucination detection to self-consistency, BERTScore, natural language inference, and claim verification.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of LLM Evaluation and Alignment, The Foundational Ideas. The section links point to the parts of the book it draws on.