mlsanityData to 7 Oct 2026

Rubrics on Trial: Evolving Rubrics from a Single Query via Synthetic Pairwise Evidence

arXiv 2607.15092 ↗First posted 16 Jul 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Rubric evolution offers a promising approach to improving the quality of rubrics generated by large language models (LLMs). Central to this process is rubric comparison, which identifies the better of two rubrics and guides the direction of evolution. However, accurate rubric comparison is difficult, which presents two challenges. (1) It should reflect downstream task performance, which is essential for assessing rubric utility but often prohibitively expensive to evaluate. (2) It should discourage unnecessary criteria, which increase verification costs and may dilute the influence of essential criteria. To address these challenges, we introduce Rubrics on Trial, a multi-agent framework that evolves rubrics by comparing synthetic response pairs. To address challenge 1, the framework compares synthetic responses that satisfy the respective rubrics, providing a proxy for downstream performance without training a separate policy for each rubric. To address challenge 2, it assesses the necessity of a candidate criterion by independently generating high-quality alternative responses that violate it and comparing them with edited versions that satisfy it. A rubric is favored when it improves response quality in both comparisons, and the resulting comparison signal is further incorporated for rubric evolution. Extensive experiments demonstrate that Rubrics on Trial improves the quality of generated rubrics and leads to better downstream task performance.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 49 of its 53 references that are in the map. First posted on arXiv 16 Jul 2026.

Builds on

49 of 53 references.

References · 49

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  4. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  5. 2024WildChat: 1M ChatGPT Interaction Logs in the Wildcited by 598
  6. 2024From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipelinecited by 484
  7. 2024RewardBench: Evaluating Reward Models for Language Modelingcited by 480
  8. 2025HealthBench: Evaluating Large Language Models Towards Improved Human Healthcited by 385
  9. 2025Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domainscited by 328
  10. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 320
  11. 2023Large Language Models are Effective Text Rankers with Pairwise Ranking Promptingcited by 302
  12. 2024JudgeBench: A Benchmark for Evaluating LLM-based Judgescited by 294
37 more
  1. 2024RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Stylecited by 168
  2. 2025RewardBench 2: Advancing Reward Model Evaluationcited by 136
  3. 2025OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignmentcited by 118
  4. 2025Checklists Are Better Than Reward Models For Aligning Language Modelscited by 115
  5. 2025Reinforcement Learning with Rubric Anchorscited by 108
  6. 2025DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Researchcited by 98
  7. 2024The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Modelscited by 89
  8. 2025DeepAgent: A General Reasoning Agent with Scalable Toolsetscited by 69
  9. 2023LLM Comparative Assessment: Zero-shot NLG Evaluation through Pairwise Comparisons using Large Language Modelscited by 66
  10. 2024TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generationcited by 62
  11. 2026Rethinking Rubric Generation for Improving LLM Judge and Reward Modeling for Open-ended Taskscited by 55
  12. 2025Chasing the Tail: Effective Rubric-based Reward Modeling for Large Language Model Post-Trainingcited by 46
  13. 2026Alternating Reinforcement Learning for Rubric-Based Reward Modeling in Non-Verifiable LLM Post-Trainingcited by 45
  14. 2026RubricHub: A Comprehensive and Highly Discriminative Rubric Dataset via Automated Coarse-to-Fine Generationcited by 42
  15. 2025Online Rubrics Elicitation from Pairwise Comparisonscited by 41
  16. 2025RocketEval: Efficient Automated LLM Evaluation via Grading Checklistcited by 37
  17. 2025Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learningcited by 32
  18. 2025LLM-Powered GUI Agents in Phone Automation: Surveying Progress and Prospectscited by 31
  19. 2026Autorubric: A Unifying Framework for Rubric-Based LLM Evaluation on Non-Verifiable Taskscited by 23
  20. 2026RubricBench: Aligning Model-Generated Rubrics with Human Standardscited by 21
  21. 2026Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generationcited by 18
  22. 2026Reward Hacking in Rubric-Based Reinforcement Learningcited by 18
  23. 2026EvoLM: Self-Evolving Language Models through Co-Evolved Discriminative Rubricscited by 12
  24. 2026CDRRM: Contrast-Driven Rubric Generation for Reliable and Interpretable Reward Modelingcited by 10
  25. 2026RubricRAG: Towards Interpretable and Reliable LLM Evaluation via Domain Knowledge Retrieval for Rubric Generationcited by 10
  26. 2026EvoRubric: Self-Evolving Rubric-Driven RL for Open-Ended Generationcited by 7
  27. 2026SibylSense: Adaptive Rubric Learning via Memory Tuning and Adversarial Probingcited by 6
  28. 2026CausalRM: Causal-Theoretic Reward Modeling for RLHF from Observational User Feedbackscited by 6
  29. 2026Qworld: Question-Specific Evaluation Criteria for LLMscited by 6
  30. 2026CausalGame: Benchmarking Causal Thinking of LLM Agents in Gamescited by 5
  31. 2026Uncertainty-Aware Reward Modeling for Stable RLHFcited by 4
  32. 2026MuSEAgent: A Multimodal Reasoning Agent with Stateful Experiencescited by 3
  33. 2026Optimal Transport for LLM Reward Modeling from Noisy Preferencecited by 3
  34. 2026RUBRIC-ARROW: Alternating Pointwise Rubric Reward Modeling for LLM Post-training in Non-verifiable Domainscited by 3
  35. 2026ImplicitRM: Unbiased Reward Modeling from Implicit Preference Data for LLM alignmentcited by 2
  36. 2026Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modelingcited by 2
  37. 2026MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Managementcited by 1

Built on by

Not cited yet.