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

Question Answering on SQuAD v2 (Utility/ASR)

1ASR Score

No Defense

-0.040.230.50.77Apr 9, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.04
10.32
2026.04
10.36
2026.04
1-
2026.04
0.970.52
2026.04
0.930.38
2026.04
0.860.56
2026.04
0.850.48
2026.04
0.78-
2026.04
0.770.66
2026.04
0.740.65
2026.04
0.71-
2026.04
0.60.74
2026.04
0.47-
2026.04
0.240.69
2026.04
0.24-
2026.04
0.15-
2026.04
0.15-
2026.04
0.090.91
2026.04
0.040.95
2026.04
0.010.95
2026.04
0.010.84
2026.04
0.010.78
2026.04
0.010.82
2026.04
0.010.99
2026.04
01
2026.04
00.99
2026.04
01
2026.04
00.99
2026.04
00.96
2026.04
0-
2026.04
00.61
2026.04
0-
2026.04
0-
2026.04
0-
2026.04
01
2026.04
0-