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

Question Answering on BoolQ (Loss and Accuracy)

0.23Loss

MeZO

0.21120.33810.4650.5919Oct 1, 2025
Updated 2d ago

Evaluation Results

MethodLinks
2025.10
0.2384
2025.10
0.2586
2025.10
0.2989
2025.10
0.3286
2025.10
0.3487
2025.10
0.3584
2025.10
0.3679
2025.10
0.4183
2025.10
0.4278
2025.10
0.575
2025.10
0.5866
2025.10
0.666
2025.10
0.6167
2025.10
0.6261
2025.10
0.6264
2025.10
0.6266
2025.10
0.6265
2025.10
0.6363
2025.10
0.6662
2025.10
0.766