Citation / Evidence Attribution Quality
The quality of whether citations or evidence attached to a claim actually support that claim and correctly point to the relevant material.
Example
A response contains several citations, but none includes a passage directly supporting its main conclusion.
Distinctions and cautions
Merely attaching a citation does not establish good attribution quality. Factuality focuses on whether a claim agrees with facts. Groundedness may encompass broader support of an output by evidence. Citation / Evidence Attribution Quality is not identical to either.
Terminology note
Specific metrics and assessment methods for citation completeness, sufficiency, and accuracy may vary depending on the use case and evaluation design.
Evidence
View evidence →
Enabling Large Language Models to Generate Text with Citations
- Title
- Enabling Large Language Models to Generate Text with Citations
- Author / organization
- Gao et al.
- Year
- 2023
- Source type / publication status
- EMNLP peer-reviewed research; published
- Supported concepts / claim context
- Citation accuracy, completeness, actual support, retrieval.
- Scope and limitations
- Dependent on ALCE benchmark/tasks; the presence of a citation does not guarantee support.
- Access / version note
- Published EMNLP.
- Recorded official URL / DOI / identifier
- DOI:
10.18653/v1/2023.emnlp-main.398(DOI resolver)
Overview of the TREC 2025 Retrieval Augmented Generation Track
- Title
- Overview of the TREC 2025 Retrieval Augmented Generation Track
- Author / organization
- NIST TREC 2025
- Year
- 2025
- Source type / publication status
- Official evaluation proceedings; published
- Supported concepts / claim context
- Grounding, completeness, attribution, retrieval/relevance, RAG evaluation.
- Scope and limitations
- Operational definitions for a shared evaluation task, not standardized definitions.
- Access / version note
- Official proceedings page.
- Recorded official URL / DOI / identifier
- Official URL / recorded link