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

Long-context Question Answering on DetectiveQA-Zh

80Accuracy

MiA-RAG

-2.57256818.86454140.3016561.738759Dec 19, 2025Jan 11, 2026Feb 3, 2026Feb 26, 2026Mar 21, 2026Apr 13, 2026May 7, 2026
Updated 26d ago

Evaluation Results

MethodLinks
2026.05
8042.7
2026.05
7842.7
2026.05
7642.7
2026.05
73.342.7
2026.05
6829.3
2026.05
56.729.3
2025.12
0.8417-
2025.12
0.8317-
2025.12
0.8267-
2025.12
0.82-
2025.12
0.8183-
2025.12
0.8167-
2025.12
0.8133-
2025.12
0.8117-
2025.12
0.8083-
2025.12
0.8017-
2025.12
0.7967-
2025.12
0.795-
2025.12
0.79-
2025.12
0.79-
2025.12
0.7883-
2025.12
0.785-
2025.12
0.785-
2025.12
0.7833-
2025.12
0.78-
2025.12
0.7717-
2025.12
0.7617-
2025.12
0.7617-
2025.12
0.76-
2025.12
0.7567-
2025.12
0.7533-
2025.12
0.7367-
2025.12
0.73-
2025.12
0.7083-
2025.12
0.705-
2025.12
0.685-
2025.12
0.6367-
2025.12
0.6033-