Accuracy
The extent to which a prediction, calculation, estimate, answer, or similar output agrees with the correct answer, a benchmark value, or an accepted reference value.
Example
A model answers ten arithmetic questions, and eight answers agree with the reference solutions.
Distinctions and cautions
How accuracy is measured depends on the task; it does not mean only classification accuracy. Whether the system expresses high confidence, or whether its confidence is well calibrated, is a different question.
Evidence
View evidence →
Artificial Intelligence Risk Management Framework (AI RMF 1.0)
- Title
- Artificial Intelligence Risk Management Framework (AI RMF 1.0)
- Author / organization
- NIST
- Year
- 2023
- Source type / publication status
- Official technical publication; published
- Supported concepts / claim context
- Reliability, Accuracy, Robustness; trustworthiness characteristics.
- Scope and limitations
- An AI-wide risk framework, not an LLM-specific taxonomy. A revision was underway as of 2026.
- Access / version note
- Treat the published edition as the historical/technical baseline.
- Recorded official URL / DOI / identifier
- DOI:
10.6028/NIST.AI.100-1(DOI resolver)
On Calibration of Modern Neural Networks
- Title
- On Calibration of Modern Neural Networks
- Author / organization
- Guo, Pleiss, Sun & Weinberger
- Year
- 2017
- Source type / publication status
- ICML peer-reviewed research; published
- Supported concepts / claim context
- Calibration; confidence versus correctness; temperature scaling.
- Scope and limitations
- Primarily studies neural networks for classification. Results cannot be directly generalized to verbal confidence in LLMs.
- Access / version note
- Published conference proceedings.
- Recorded official URL / DOI / identifier
- Official URL / recorded link
FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation
- Title
- FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation
- Author / organization
- Min et al.
- Year
- 2023
- Source type / publication status
- EMNLP peer-reviewed research; published
- Supported concepts / claim context
- Factuality, Verification, claim-level support.
- Scope and limitations
- Mainly biographies and long-form generation; depends on the quality of reference knowledge sources.
- Access / version note
- Published EMNLP.
- Recorded official URL / DOI / identifier
- DOI:
10.18653/v1/2023.emnlp-main.741(DOI resolver)
Evaluating large language models for accuracy incentivizes hallucinations
- Title
- Evaluating large language models for accuracy incentivizes hallucinations
- Author / organization
- Kalai et al.
- Year
- 2026
- Source type / publication status
- Nature peer-reviewed research; published
- Supported concepts / claim context
- Accuracy, Abstention, Hallucination; evaluation incentives.
- Scope and limitations
- Formal model and selected frontier-model case study. Utility depends on the use case.
- Access / version note
- Published Nature article.
- Recorded official URL / DOI / identifier
- DOI:
10.1038/s41586-026-10549-w(DOI resolver)