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

Document Question Answering on LongDocURL

71.4Accuracy (All)

GPT-4o-200b-128

0.36818.80937.2555.691Oct 8, 2025Nov 7, 2025Dec 8, 2025Jan 8, 2026Feb 7, 2026Mar 10, 2026Apr 10, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2025.10
71.474.668.6
2025.10
65.972.459.9
2025.10
65.270.960
2025.10
63.466.860.3
2025.10
63.165.960.6
2025.10
62.968.458
2025.10
62.270.255.2
2025.10
5968.451
2026.04
56.4--
2026.04
53.6--
2026.04
52.8--
2026.04
52.7--
2025.10
50.755.146.8
2026.04
50.3--
2025.10
48.949.347.6
2025.10
47.948.347.6
2025.10
47.751.843.8
2025.10
47.648.546.8
2025.10
46.948.445.5
2025.10
46.851.541.1
2025.10
45.749.942
2025.10
45.548.842.3
2025.10
44.348.440.4
2025.10
4345.840.1
2025.10
42.745.740
2026.04
39.8--
2026.04
11--
2025.10
66.65.3
2025.10
4.55.43.8
2025.10
3.13.92.4