Revision History
TruthfulQA · 13 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Update: regenerate results chart from the fact-checked results table
Version 13 · Nov 2, 2026, 01:18 PM
- Correction: relabel the TruthfulQA results table column and chart from %True to Score, since most rows are multiple-choice accuracy rather than generation %True
Version 12 · Nov 2, 2026, 01:18 PM
- Correction: Corrected the category table (wrong counts, two missing categories and two invented ones), replaced example questions that are not in the dataset, fixed a misattributed leaderboard table (self-reported LLM-Stats scores labeled as Open LLM Leaderboard v1 MC2), the GPT-3.5 Turbo and Llama 2 70B chat scores, and a false claim that GPT-judge checkpoints are provided (plus 28 minor fixes)
Version 11 · Nov 2, 2026, 01:17 PM
- Improve internal links (2026-10-13)
Version 8 · Oct 13, 2026, 10:15 AM
- Added 6 contextual internal links
Version 7 · Jul 23, 2026, 03:27 PM