Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Error Recognition on Math500
Loading...
57.1
Top-1 Accuracy
CORRECT
8.324
20.987
33.65
46.313
Sep 28, 2025
Top-1 Accuracy
Top-3 Accuracy
Top-5 Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Top-1 Accuracy
Top-3 Accuracy
Top-5 Accuracy
CORRECT
Synthesized by=GPT-4o-...
2025.09
57.1
87.8
95.9
CORRECT
Synthesized by=GPT-5-Nano
2025.09
57.1
87.8
95.9
Baseline
Synthesized by=GPT-5-Nano
2025.09
41.8
57.1
64.3
LLM-as-a-Judge
Synthesized by=GPT-4o-...
2025.09
10.2
23.4
35.6
Feedback
Search any
task
Search any
task