Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Error Detection on 74-paper benchmark (full)
Loading...
55.4
Recall
OpenAIReview
10.368
22.059
33.75
45.441
Jun 18, 2026
Recall
Updated 1mo ago
Evaluation Results
Method
Method
Links
Recall
OpenAIReview
Backend=DeepSeek-V4-Flash
2026.06
55.4
OpenAIReview
Backend=Grok-4.1-Fast
2026.06
49.3
OpenAIReview
Backend=Qwen3.6-35B-A3B
2026.06
46.2
zero-shot
Backend=Grok-4.1-Fast
2026.06
33.5
zero-shot
Backend=Qwen3.6-35B-A3B
2026.06
31.7
OpenAIReview
Backend=Gemini-3.1-Fla...
2026.06
30.9
zero-shot
Backend=DeepSeek-V4-Flash
2026.06
29.4
coarse
Backend=DeepSeek-V4-Flash
2026.06
21.2
coarse
Backend=Qwen3.6-35B-A3B
2026.06
15.9
coarse
Backend=Grok-4.1-Fast
2026.06
14.8
zero-shot
Backend=Gemini-3.1-Fla...
2026.06
14.3
coarse
Backend=Gemini-3.1-Fla...
2026.06
12.1
Feedback
Search any
task
Search any
task