Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Software Engineering Skill Evaluation on SWE-Skills-Bench implementing-agent-modes
Loading...
30
Pass Rate
Full Skill Text
19.6
22.3
25
27.7
Jun 15, 2026
Pass Rate
Spearman Correlation (ρ)
CNG Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pass Rate
Spearman Correlation (ρ)
CNG Score
Full Skill Text
Setting=Full Skill Tex...
2026.06
30
88.53
0.11
S2L
Setting=Our method
2026.06
30
21.91
0.46
Vanilla LLM
Setting=Vanilla baseline
2026.06
20
-
-
Feedback
Search any
task
Search any
task