Revision History
HotpotQA · 5 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.
- Correction: Corrected the human-performance figures (the 68.99/52.37 scores were the baseline model's, humans scored 91.40 Answer F1), fixed wrong or mislabeled scores for GoldEn Retriever, Cognitive Graph, HGN, the Asai path retriever, Longformer, MDR, ReAct and IRCoT, and removed fabricated claims that GPT-3, Toolformer, Self-RAG, Self-Ask, RAGAS and RGB evaluated on HotpotQA and an invented worker-pay figure (plus 10 minor fixes)
Version 5 · Nov 4, 2026, 09:48 PM
- Improve internal links (2026-10-13)
Version 4 · Oct 13, 2026, 10:20 AM
- Added 6 contextual internal links
Version 3 · Jul 23, 2026, 03:29 PM