Deep Research as Rubric for Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mei, Wangyi, Gu, Zhouhong, Bai, Zhenhan, Cai, Yin, Zhang, Lefan, Ding, Zhenxin, Chen, Bo, Gao, Yan, Wu, Yi, Hu, Yao, Liang, Jiaqing, Yang, Deqing
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917551203680256
author Mei, Wangyi
Gu, Zhouhong
Bai, Zhenhan
Cai, Yin
Zhang, Lefan
Ding, Zhenxin
Chen, Bo
Gao, Yan
Wu, Yi
Hu, Yao
Liang, Jiaqing
Yang, Deqing
author_facet Mei, Wangyi
Gu, Zhouhong
Bai, Zhenhan
Cai, Yin
Zhang, Lefan
Ding, Zhenxin
Chen, Bo
Gao, Yan
Wu, Yi
Hu, Yao
Liang, Jiaqing
Yang, Deqing
contents Open-ended reasoning and long-form generation tasks lack reliable automatic verification signals for reward-based policy optimization. Rubrics offer a promising alternative, but existing approaches treat them as given artifacts -- either hand-crafted or prompt-generated -- and often miss the task-specific, knowledge-intensive dimensions that matter most, distorting the reward signal. Our key observation is that rubric construction is itself a research problem: identifying what makes a response correct or insightful requires discovering and synthesizing external knowledge. We propose Deep Research as Rubric (DR-rubric), a two-stage framework for constructing such rubrics. Stage I elicits domain facts, structural constraints, and failure modes through iterative multi-turn agentic search; Stage II distills this evidence into atomic, independently verifiable constraints for GRPO-based policy optimization. Because the model under training can serve as its own rubric generator, DR-rubric-8B supports bootstrap rubric generation without frontier-model assistance. We evaluate on 6 benchmarks spanning agentic research and expert reasoning. Experiments show that DR-Rubric achieves strong competitive performance with only 1K -- 3K training instances, where GPT-5-generated rubrics particularly benefit breadth coverage on agentic tasks, Gemini-generated rubrics yield the most balanced performance across agentic and expert reasoning tasks, and bootstrap rubrics exhibit a specialization-to-rebalancing evolution achieving the best overall performance at the third iteration. Results demonstrate that reframing rubric construction from static evaluation templates into an evidence-driven research process yields more scalable, fine-grained reward signals for open-ended tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2606_01091
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Deep Research as Rubric for Reinforcement Learning
Mei, Wangyi
Gu, Zhouhong
Bai, Zhenhan
Cai, Yin
Zhang, Lefan
Ding, Zhenxin
Chen, Bo
Gao, Yan
Wu, Yi
Hu, Yao
Liang, Jiaqing
Yang, Deqing
Computation and Language
Open-ended reasoning and long-form generation tasks lack reliable automatic verification signals for reward-based policy optimization. Rubrics offer a promising alternative, but existing approaches treat them as given artifacts -- either hand-crafted or prompt-generated -- and often miss the task-specific, knowledge-intensive dimensions that matter most, distorting the reward signal. Our key observation is that rubric construction is itself a research problem: identifying what makes a response correct or insightful requires discovering and synthesizing external knowledge. We propose Deep Research as Rubric (DR-rubric), a two-stage framework for constructing such rubrics. Stage I elicits domain facts, structural constraints, and failure modes through iterative multi-turn agentic search; Stage II distills this evidence into atomic, independently verifiable constraints for GRPO-based policy optimization. Because the model under training can serve as its own rubric generator, DR-rubric-8B supports bootstrap rubric generation without frontier-model assistance. We evaluate on 6 benchmarks spanning agentic research and expert reasoning. Experiments show that DR-Rubric achieves strong competitive performance with only 1K -- 3K training instances, where GPT-5-generated rubrics particularly benefit breadth coverage on agentic tasks, Gemini-generated rubrics yield the most balanced performance across agentic and expert reasoning tasks, and bootstrap rubrics exhibit a specialization-to-rebalancing evolution achieving the best overall performance at the third iteration. Results demonstrate that reframing rubric construction from static evaluation templates into an evidence-driven research process yields more scalable, fine-grained reward signals for open-ended tasks.
title Deep Research as Rubric for Reinforcement Learning
topic Computation and Language
url https://arxiv.org/abs/2606.01091