Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Semantic Correctness Evaluation on RepoProver
Loading...
82
Faithfulness
NL vs. Lean
68.48
71.99
75.5
79.01
Jun 12, 2026
Faithfulness
Claim Strength: Stronger (%)
Claim Strength: Different (%)
Overall Agreement (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Faithfulness
Claim Strength: Stronger (%)
Claim Strength: Different (%)
Overall Agreement (%)
NL vs. Lean
Evaluation Protocol=di...
2026.06
82
1
18
78.4
NL vs. NLb
Evaluation Protocol=ro...
2026.06
69
11
19
78.4
Feedback
Search any
task
Search any
task
