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

Long-context Question Answering on MFQA En

29.33SubEM

LongReward-PO

16.153219.574122.99526.4159Aug 19, 2025
Updated 9d ago

Evaluation Results

MethodLinks
2025.08
29.3344.89
2025.08
2845.12
2025.08
27.3350.29
2025.08
27.3341.08
2025.08
26.6747.78
2025.08
26.6748.19
2025.08
2651.26
2025.08
25.3348.87
2025.08
25.3348.07
2025.08
24.6746.83
2025.08
24.6749.18
2025.08
24.6738.08
2025.08
24.6645.88
2025.08
2445.48
2025.08
23.3347.07
2025.08
23.3347.73
2025.08
23.3347.32
2025.08
23.3347.57
2025.08
23.3349.96
2025.08
23.3347.92
2025.08
23.3349.76
2025.08
22.6747.99
2025.08
22.6748.56
2025.08
22.6746.74
2025.08
2248.49
2025.08
21.3345.03
2025.08
21.3344.06
2025.08
20.6746.25
2025.08
20.6746.92
2025.08
20.6648.1
2025.08
2044.7
2025.08
19.3344.12
2025.08
1944.39
2025.08
18.6644.34
2025.08
17.3338.01
2025.08
16.6640.67