← Back to all news
arXiv AI By Sumeet Ramesh Motwani, Tejas Krishnan, Alistair Letcher, Charles London, Suhaas Bhat, Huitian Jiao, Philip Torr, Riashat Islam, Christopher Summerfield, Jakob Foerster, Christian Schroeder de Witt, Qilong Gu, Shital Shah

Rubric Curriculum RL: Exploiting the Generation-Verification Gap in Non-Verifiable Domains

Read the original on arXiv AI →

Rubric Curriculum RL (RcRL) is a reinforcement learning approach that leverages the generation‑verification gap in non‑verifiable domains by using pairwise preferences over rollouts and a curriculum over rubric criteria. Unlike traditional methods that rely on absolute scoring and stationary judges, RcRL provides a more discriminative, non‑stationary signal, allowing policies to maintain output entropy and continue improving beyond 1,000 training steps. In experiments on creative writing and HealthBench, RcRL outperforms baseline models, achieving a 70.5% human win rate and significant accuracy gains over the base model.

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.

Related stories

arXiv Machine Learning
4d ago

Step-wise Rubric Rewards for LLM Reasoning

The paper introduces Step-wise Rubrics as Rewards (SRaR), a reinforcement learning framework that assigns rubric-based scores to individual reasoning steps in large language models. By normalizing step scores and combining them with outcome rewards through a decoupled advantage estimator, SRaR preserves criterion-level structure and mitigates issues like rewarding flawed steps or encouraging endless self‑correction. Experiments on 16K problems and six math benchmarks show that SRaR outperforms prior rubric methods, improving accuracy, increasing faithful reasoning rates, and reducing self‑correction loops.

By Weichu Xie, Haozhe Zhao, Wenpu Liu, Yongfu Zhu, Liang Chen, Minghao Ye, Zirong Chen, Yuqi Xu, Shuai Dong, Ziyue Wang, Xinbo Xu, Kean Shi, Ruoyu Wu, Xiaoying Zhang, Wenqi Shao, Baobao Chang, Nan Duan, Jiaqi Wang
llmsreinforcement-learningbenchmarks
More like this →
arXiv AI
Aug 5

Rubrics as Privileged Information for Open-Ended Generation

arXiv:2608. 02948v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD), where a single model acts as both student and teacher with different contexts, has shown promise in verifiable domains like math, where hard privileged information (PI) in the form of ground-truth answers structurally constrains valid continuations.

By Deepika Bablani, Ajay Gupta, Wanming Chen
llmsreinforcement-learningefficiencybenchmarks
More like this →
arXiv AI
Jun 3

QUBRIC: Co-Designing Queries and Rubrics for RL Beyond Verifiable Rewards

arXiv:2606. 03968v1 Announce Type: cross Abstract: Rubric-based RL is a promising route for extending reinforcement learning beyond verifiable rewards, yet existing methods optimize rubrics while treating the query distribution as fixed.

By Rongzhi Zhang, Rui Feng, Zhihan Zhang, Jingfeng Yang, Qingyu Yin, Xin Liu, Zixuan Zhang, Priyanka Nigam, Bing Yin, Tuo Zhao, Chao Zhang
reinforcement-learningfine-tuningbenchmarks
More like this →