arXiv AI October 16, 2026 By Licheng Pan, Haocheng Yang, Haoxuan Li, Yuzhou Chen, Yichen Sun, Yuan Lu, Xiaoxi Li, Hao Wang, Zhixuan Chu, Kui Ren

Rubric Judgments Lack Uncertainty: Conformal Weighting for LLM Evaluation

Read the original on arXiv AI →

Rubric-based evaluation of large language models (LLMs) often assigns binary judgments to multiple criteria, but the certainty of these judgments varies widely. Existing aggregation methods treat all criteria as equally certain, potentially misleading reinforcement learning signals. The authors propose Uncertainty-aware Rubric Weighting (URW), a training-free framework that uses conformal calibration to quantify and weight criterion-level uncertainty, improving preference accuracy and downstream RL alignment.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Computation and Language

Sep 1

Small Language Models as Judges for Rubric-Based Reinforcement Learning

arXiv:2608.30005v1 Announce Type: new Abstract: Rubric-based reinforcement learning extends RL beyond tasks with exact answers or rule-based verifiers by scoring responses against instance-specific c...

By Fengyu Xie, Yilun Zhao, Bingsen Chen, Arman Cohan, Chen Zhao

llms reinforcement-learning

arXiv Computation and Language

4d ago

Not All Criteria Matter: Importance-Aware Criteria Weighting for Rubric-Based LLM Evaluation

arXiv:2610.19915v1 Announce Type: new Abstract: Rubric-based evaluation assesses the responses of large language models (LLMs) by aggregating the verification results of explicit criteria, yet not al...

By Haocheng Yang, Licheng Pan, Yiming Dong, Xiaoxi Li, Yuan Lu, Zhenhao Chen, Jiajun Fan, Shuai Shao, Chunyuan Zheng, Hao Wang

llms reinforcement-learning

Hugging Face Trending Papers

Jul 2

Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modeling

Reliable reward and preference signals are critical for evaluating and optimizing large language models on open-ended tasks. Rubric-based judges offer a transparent way to decompose such judgments into explicit evaluation criteria, but existing annotation-free rubric generators typically rely on a single generic evaluator.

llms reinforcement-learning benchmarks

arXiv Computation and Language

Oct 8

Judging in Latent Space: Efficient Generative Reward Modeling via Semantics-Preserving Compression

arXiv:2610.09788v1 Announce Type: new Abstract: Reward modeling often requires jointly representing and reasoning over multiple evaluation criteria, yet verbalizing this process token by token can in...

By Mingqing Yuan (Soochow University), Xiaobo Liang (Soochow University), Junwei Yang (University of Cambridge), Ziwei Chen (Chalmers University of Technology), Zeren Zhang (Peking University), Hejin Wang (Tsinghua University), Yubin Wang (The Hong Kong University of Science,Technology), Juntao Li (Soochow University)

rag agents diffusion reinforcement-learning fine-tuning benchmarks

arXiv Machine Learning

Jul 3

Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modeling

arXiv:2607. 01830v1 Announce Type: new Abstract: Reliable reward and preference signals are critical for evaluating and optimizing large language models on open-ended tasks.

By Dazhi Fu, Jiuding Yang, Yiwen Guo, Jicong Fan

llms reinforcement-learning benchmarks

Hugging Face Trending Papers

5d ago

Rubrics from the Task, Verdicts from Evidence: Pointwise Rubric Rewards with Tool-Augmented Judges

The paper introduces ReVeR, a pointwise rubric reward model that derives hard rules and graded principles directly from instructions, avoiding response-induced bias. It employs tool-augmented judges and label‑blind distillation to create ReVeR‑Traces, a large dataset of rubric reasoning and judging trajectories. Experiments show ReVeR surpasses pairwise rubric models in preference accuracy, especially for instruction following, and reveal that higher preference accuracy does not always lead to better downstream policies.

reinforcement-learning efficiency safety

One email a morning, machine-written

Related stories

Small Language Models as Judges for Rubric-Based Reinforcement Learning

Not All Criteria Matter: Importance-Aware Criteria Weighting for Rubric-Based LLM Evaluation

Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modeling

Judging in Latent Space: Efficient Generative Reward Modeling via Semantics-Preserving Compression

Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modeling

Rubrics from the Task, Verdicts from Evidence: Pointwise Rubric Rewards with Tool-Augmented Judges