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

• URL: http://arxiv.org/abs/2606.08077v1
• Date: Sat, 06 Jun 2026 09:55:22 GMT
• Title: Support Vector Rubrics: Closing the Gap Between Self-Generated and Human Rubrics
• Abstract summary: We introduce SVR (Support Vectors), a framework that recasts rubrics as max-margin boundary learning over preference data.<n> SVR mines contrastive features from preference pairs into a bank, learns a prompt-conditioned selector together with global weights, and iteratively refines bank through support-pair selection and adversarial probing of hard negatives.<n>OnBench, SVR narrows the gap to human references from 24.1 to 0.3 points and outperforms strong self-rubrics, and the learned bank transfers across judges without retraining.
• Score: 33.21526869949403
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Rolling With Resistance: Preference-Optimized LLM Counselors Can Trade Goal Persistence for Relational Attunement in Motivational Interviewing [0.6372261626436676]
  In Motivational Interviewing (MI), a client's sustain talk (arguments for the status quo) calls for the counselor to roll with resistance, a move that can fail in two ways.<n>We introduce a two-axis evaluation of counselor responses of Motivational Interviewing Treatment Integrity (MITI) code, Goal Persistence (GP) and Attunement (RA)<n>We ask whether penalizing one failure through preference optimization teaches rolling resistance provokes its opposite.
  arXiv  Detail & Related papers  (2026-07-30T20:16:03Z)
• Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modeling [39.325422496065435]
  Multi-Role Generation (MRRG) is a training-free and reference-free framework that elicits evaluation criteria from multiple roles.<n>MRRG can be used both to validate pairwise preferences and to provide rewards for RLO-style Reinforcement Learning with Verifiable Rewards (RLVR)
  arXiv  Detail & Related papers  (2026-07-02T07:50:38Z)
• PerceptionRubrics: Calibrating Multimodal Evaluation to Human Perception [75.70273182182397]
  PerceptionRubrics is a rubric-based evaluation framework.<n>It addresses the gap between saturated benchmark scores and real-world brittleness.<n>It pairs 1,038 information-dense images with over 10,000 instance-specific rubrics.
  arXiv  Detail & Related papers  (2026-06-26T17:59:15Z)
• Rethinking Reward Supervision: Rubric-Conditioned Self-Distillation [60.55792673956761]
  We propose textbfRubric-Conditioned Self-Distillation, a framework that incorporates rubrics as structured, fine-grained feedback for on-policy self-distillation.<n>We evaluate on a diverse suite of science reasoning benchmarks and results show that rubric-conditioned self-distillation effectively converts rubric-level criteria into token-level guidance.
  arXiv  Detail & Related papers  (2026-06-17T17:54:04Z)
• ARBOR: Online Process Rewards via a Reusable Rubric Buffer for Search Agents [48.80766702702854]
  LLM-based search agents are trained predominantly with outcome-only reward, leaving the search process itself unsupervised.<n>This signal degenerates on outcome-homogeneous groups where all sampled trajectories share the same correctness, yielding zero within-group advantage and no gradient.<n>We propose ARBOR (Adaptive Buffer for Online Reward), a reusable process-reward framework that maintains a rubric memory shared across queries.
  arXiv  Detail & Related papers  (2026-06-02T06:58:54Z)
• CriterAlign: Criterion-Centric Rationale Alignment for Code Preference Judging [95.02210956333374]
  We propose a criterion-centric framework that adapts rubric-based judging to pairwise preference evaluation.<n>On BigCodeReward, CriterAlign improves a Qwen2.5-VL-32B monolithic judge from 60.4% to 66.3% accuracy.
  arXiv  Detail & Related papers  (2026-05-19T10:59:19Z)
• GRASP: Deterministic argument ranking in interaction graphs [65.66879897437157]
  We show that holistic judging suffers from substantial inter-model disagreement.<n>This instability arises from collapsing a debate's complex interaction structure into a single opaque score.<n>We propose GRASP, a deterministic framework that aggregates stable local interaction judgments into a global ranking.
  arXiv  Detail & Related papers  (2026-05-18T21:49:02Z)
• Auto-Rubric as Reward: From Implicit Preferences to Explicit Multimodal Generative Criteria [17.272139541614383]
  Auto-Rubric as Reward (ARR) is a framework that reframes reward modeling from implicit weight optimization to explicit, criteria-based decomposition.<n>ARR externalizes a VLM's preference knowledge as prompt-specific gradients, translating holistic intent into independently verifiable quality dimensions.<n>ARR-RPO outperforms pairwise reward models and VLM judges on text-to-image generation and image editing benchmarks.
  arXiv  Detail & Related papers  (2026-05-08T18:05:27Z)
• CDRRM: Contrast-Driven Rubric Generation for Reliable and Interpretable Reward Modeling [61.75914342638658]
  CDRRM (Contrast-Driven Reward Model) is a framework for high-quality rubric generation and guided preference judgment.<n>This work offers a scalable, interpretable, and data-efficient path for reward modeling.
  arXiv  Detail & Related papers  (2026-03-09T07:15:23Z)
• R-Align: Enhancing Generative Reward Models through Rationale-Centric Meta-Judging [69.96389360650072]
  We show that reasoning fidelity is highly predictive of downstream RLHF outcomes, beyond standard label accuracy.<n>We propose Rationale-Centric Alignment, R-Align, which augments training with gold judgments and explicitly supervises rationale alignment.
  arXiv  Detail & Related papers  (2026-02-06T15:17:11Z)
• Rethinking Rubric Generation for Improving LLM Judge and Reward Modeling for Open-ended Tasks [17.117706938140078]
  We propose RRD, a principled framework for rubric refinement built on a decompose-filter cycle.<n> RRD decomposes coarse rubrics into fine-grained, discriminative criteria, expanding coverage while sharpening separation between responses.<n>It delivers large, consistent gains across both evaluation and training.
  arXiv  Detail & Related papers  (2026-02-04T23:16:09Z)
• BoRP: Bootstrapped Regression Probing for Scalable and Human-Aligned LLM Evaluation [13.561789180344533]
  We introduce BoRP, a scalable framework for high-fidelity satisfaction evaluation.<n>Experiments on industrial datasets show that BoRP significantly outperforms generative baselines.<n>BoRP reduces inference costs by orders of magnitude, enabling full-scale monitoring and highly sensitive A/B testing via CUPED.
  arXiv  Detail & Related papers  (2026-01-26T08:20:02Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment [38.1645520104553]
  We introduce OpenRubrics, a large-scale collection of (prompt,explicit) pairs for training rubric-generation and rubric-based reward models.<n>To elicit discriminative and comprehensive evaluation signals, we introduce Contrastive Generation (CRG), which derives both hard rules (implicit qualities) and principles (implicit qualities) by contrasting preferred and rejected responses.<n>Our results show that rubrics provide scalable alignment signals that narrow the gap between costly human evaluation and automated reward modeling.
  arXiv  Detail & Related papers  (2025-10-09T03:31:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.