Pointwise evaluation on BIGGEN

0.584Spearman Corr

Human-crafted (existing) Rubrics

Updated 2d ago

Evaluation Results

Method
Human-crafted (existing) Rubrics 2026.05	0.584	0.598
Human-crafted (existing) Rubrics 2026.05	0.583	0.609
Human-crafted (existing) Rubrics 2026.05	0.552	0.594
Annotation-free Preference Learning for Rubric Generator 2026.05	0.51	0.521
Annotation-free Preference Learning for Rubric Generator 2026.05	0.49	0.504
Training-free Rubric Generation 2026.05	0.477	0.496
Training-free Rubric Generation 2026.05	0.477	0.496
Human-crafted (existing) Rubrics 2026.05	0.476	0.494
Training-free Rubric Generation 2026.05	0.461	0.453
DnA-Eval 2026.05	0.454	0.489
Training-free Rubric Generation 2026.05	0.446	0.462
Training-free Rubric Generation 2026.05	0.446	0.462
Annotation-free Preference Learning for Rubric Generator 2026.05	0.445	0.452
Annotation-free Preference Learning for Rubric Generator 2026.05	0.441	0.461
Training-free Rubric Generation 2026.05	0.44	0.455
Annotation-free Preference Learning for Rubric Generator 2026.05	0.44	0.447
Annotation-free Preference Learning for Rubric Generator 2026.05	0.439	0.459
Human-crafted (existing) Rubrics 2026.05	0.436	0.45
DnA-Eval 2026.05	0.431	0.426
CheckEval 2026.05	0.424	0.455
Training-free Rubric Generation 2026.05	0.42	0.434
Training-free Rubric Generation 2026.05	0.42	0.434
Human-crafted (existing) Rubrics 2026.05	0.384	0.382
Training-free Rubric Generation 2026.05	0.378	0.361
Prometheus 8x7B 2026.05	0.366	0.401
CheckEval 2026.05	0.343	0.392
DnA-Eval 2026.05	0.337	0.353
RubricHub 2026.05	0.332	0.383
RubricHub 2026.05	0.325	0.336
CheckEval 2026.05	0.318	0.361
RubricHub 2026.05	0.318	0.361
Prometheus 8x7B 2026.05	0.28	0.307