Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Correlation with Human Likert Scores on FinePhyEval
Loading...
0.478
Pearson's r
PQSG
0.41456
0.43103
0.4475
0.46397
Jun 24, 2026
Pearson's r
Kendall's τ
Spearman's ρ
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pearson's r
Kendall's τ
Spearman's ρ
PQSG
LLM Backbone=GPT-5.5
2026.06
0.478
0.336
0.456
PQSG
LLM Backbone=Gemini-2....
2026.06
0.467
0.306
0.406
PQSG
LLM Backbone=GPT-5
2026.06
0.43
0.263
0.366
PQSG
LLM Backbone=GPT-5.4
2026.06
0.429
0.274
0.369
PQSG
LLM Backbone=Gemini-2....
2026.06
0.417
0.259
0.354
Feedback
Search any
task
Search any
task
