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

Question Answering on HotpotQA (ATR, ASR, Accuracy)

91.3Accuracy

CleanRAG

-3.65220.99945.6570.301Aug 4, 2025Sep 13, 2025Oct 23, 2025Dec 2, 2025Jan 11, 2026Feb 20, 2026Apr 2, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2025.08
91.308
2025.08
90.809
2025.08
90.709
2025.08
90.51.59.1
2025.08
90.31.59
2025.08
89.91.59.4
2025.08
88.108.2
2025.08
8709
2025.08
8209.1
2025.08
66.5031
2026.04
55.71--
2026.04
54.12--
2025.08
54.11.533
2026.04
53.89--
2025.08
53.6038
2026.04
53.25--
2026.04
52.45--
2026.04
52.25--
2026.04
52.04--
2026.04
51.66--
2026.04
50.62--
2026.04
50.46--
2026.04
49.73--
2026.04
49.29--
2026.04
48.54--
2026.04
48.39--
2026.04
48.32--
2026.04
45.44--
2026.04
45.33--
2026.04
45.27--
2026.04
43.84--
2026.04
41.46--
2026.04
40.84--
2025.08
4.210094
2025.08
0.110098
2025.08
010090
2025.08
010093