Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Question Answering on MathQA (Robustness Metrics)
Loading...
46.72
Clean Accuracy
AGLS
4.3712
15.3656
26.36
37.3544
May 26, 2025
Clean Accuracy
Attack Accuracy
Adversarial Success Rate
Average Time
Updated 1mo ago
Evaluation Results
Method
Method
Links
Clean Accuracy
Attack Accuracy
Adversarial Success Rate
Average Time
AGLS
Model=gemma2-2B
2025.05
46.72
34.79
25.54
0.38
AGLS
Model=llama3.3-70B
2025.05
32.33
23.67
26.79
1.45
AGLS
Model=llama3.1-70B
2025.05
13.67
11.33
17.12
0.14
AGLS
Model=qwen2.5-0.5B
2025.05
6
2.33
61.17
1.81
Feedback
Search any
task
Search any
task
