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

Question Classification on TREC LongBench

68.5Accuracy

Baseline

-0.6617.29535.2553.205Feb 26, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.02
68.5
2026.02
68
2026.02
68
2026.02
67.5
2026.02
67.5
2026.02
66.5
2026.02
64.5
2026.02
64.5
2026.02
64.5
2026.02
64.5
2026.02
64
2026.02
62
2026.02
39.5
2026.02
38
2026.02
37.5
2026.02
37
2026.02
34.5
2026.02
32.5
2026.02
10.5
2026.02
10.03
2026.02
9.1
2026.02
9.03
2026.02
8.53
2026.02
8.1
2026.02
4.75
2026.02
4.5
2026.02
3.5
2026.02
3.5
2026.02
2.75
2026.02
2