Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Support Vector Rubrics: Closing the Gap Between Self-Generated and Human Rubrics

About

Rubric-based evaluation is a promising paradigm for judging large language model (LLM) outputs, yet self-generated rubrics lag human-annotated criteria on hard instances. We argue this discriminative gap reflects an objective mismatch: self-generated rubrics describe good responses, whereas effective criteria must discriminate between close candidates. To close this gap, we introduce SVR (Support Vector Rubrics), a framework that recasts rubric construction as max-margin boundary learning over preference data. SVR mines contrastive features from preference pairs into a rubric bank, learns a prompt-conditioned selector together with global rubric weights, and iteratively refines the bank through support-pair selection and adversarial probing of hard negatives. At inference, given only the prompt, SVR retrieves the top-rubrics from the bank and scores responses. On RubricBench, SVR narrows the gap to human reference rubrics from 24.1 to 0.3 points and outperforms strong self-rubric and judge baselines, and the learned bank transfers across judges without retraining. On RewardBench 1&2, and RM-Bench, it remains competitive with dedicated reward models, demonstrating broader reward modeling capability. Overall, boundary-defining rubrics offer a principled route to closing the discriminative gap in LLM evaluation.

Mengyuan Sun, Yu Li, Zhuohao Yu, Shikun Zhang, Wei Ye• 2026

Related benchmarks

TaskDatasetResultRank
Reward ModelingRewardBench
Safety Score82.8
284
Reward ModelingRM-Bench--
137
Reward ModelingRewardBench 2
Precise IF Score81.9
90
Pairwise AccuracyRubricBench 1.0 (test)
Pairwise Accuracy (Chat)81.8
13
Reward ModelingRewardBench and RM-Bench Aggregate 2
Average Score87.2
6
Showing 5 of 5 rows

Other info

Follow for update