Claim · 2021-09-08 · publication · Stated in source · Supported

The paper proposes a benchmark to measure whether a language model is truthful in generating answers to questions.

One recorded claim. The claim panel gives its qualification, the source passages with their place in each source, and how it was checked.

What it connects

TruthfulQA: Measuring How Models Mimic Human Falsehoods → states its contribution → The paper proposes a benchmark to measure whether a language model is truthful in generating answers to questions.

Where it appears

No collection page holds this claim. It is listed with its records in the explorer.

Claim and evidence

TruthfulQA: Measuring How Models Mimic Human Falsehoodsstates its contribution ↓The paper proposes a benchmark to measure whether a language model is truthful in generating answers to questions.

Machine-checked, not reviewed by an editorStated in sourceSupported

Supporting evidence: primary · 1 source

Date: 2021-09-08 · publication Records dated 2021 →

Scope and limitsThe paper's own description of its contribution; not an assessment of it.

The evidence 01

primary sourceSupports the claim

TruthfulQA: Measuring How Models Mimic Human Falsehoods

arXiv · 2021-09-08

Cite the original source

Cite the original document, not Selekoa. Check these details against it.

Source excerpt

“We propose a benchmark to measure whether a language model is truthful in generating answers to questions.”

Location in source arXiv v1, PDF p. 1, abstract

Machine-checked · 2026-10-31

Source details and check notes

Authors: Stephanie Lin; Jacob Hilton; Owain Evans
Version: arXiv:2109.07958v1, submitted 8 September 2021 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 0223117f1623e3a079056dfa6ed517a6b2fcd7141b208f395f539f33e200bdbf). The author list is the byline as printed on the first page.
Version date: 2021-09-08
Canonical address: Open the canonical source ↗
Accessed: 2026-10-28
Language: English (en)
Attribution: Stephanie Lin, Jacob Hilton, Owain Evans, “TruthfulQA: Measuring How Models Mimic Human Falsehoods”, 2021-09-08.
Reuse: No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution.
Check notes: Candidate pack passage, key papers (research/key-papers). The quotation (17 words) matched the text retrieved for the candidate pack on 2026-10-28 (research/key-papers/verification.json). Matching ignored differences of whitespace, case, quotation marks, dashes and line-break hyphens, so the source’s typography can differ. No human editorial review.

Status and meaning

Editorial status. Not reviewed by an editor. An exploratory record that no editor has accepted. This status does not say whether a person has read any linked passage. Each evidence record above shows how it was checked.

Relation. A paper or other publication describes its own contribution, quoted from the expression the record names, usually its abstract or introduction. The record is the document’s own description, not an assessment: a claim of priority or novelty it contains is the document’s claim, flagged as such, and says nothing about earlier or later work.

Date. Date of the arXiv version quoted, submitted 8 September 2021 (UTC); later versions may word it differently.

Record ID kp-states-truthfulqa-measuring-how-models-mimic-2021-1