Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Single-hop QA on NQ (test)

42Token-level F1

RAG-Critic

18.91224.90630.936.894Apr 16, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.04
42
2026.04
40.3
2026.04
40.2
2026.04
38.8
2026.04
38.4
2026.04
38.3
2026.04
32.3
2026.04
22.3
2026.04
19.8