List your company

Rubric Generation on OSS EVAL 300

1Spearman's RhoGold
0.304240.484870.66550.84613Mar 21, 2026

Evaluation results

MethodLinks
GoldRubric Source=Gold2026.03110.3660
RubricRAG (nothink)Rubric Source=RubricRAG (nothink), reasoning=disabled2026.030.5450.4780.4080.042
RubricRAG (think)Rubric Source=RubricRAG (think), reasoning=enabled2026.030.4950.4150.3740.008
Few-ShotRubric Source=Few-Shot2026.030.4660.3680.3970.031
SFTRubric Source=SFT2026.030.4570.3090.287-0.079
Zero-ShotRubric Source=Zero-Shot2026.030.4260.33-0.035-0.401
GRPORubric Source=GRPO2026.030.3310.2230.134-0.232