Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918534179717120 |
|---|---|
| author | Lv, Changze Zhou, Jie Zhao, Wentao Xu, Jingwen Dou, Shihan Huang, Zisu Tian, Muzhao Wang, Xiaohua Liu, Yang Zhou, Pluto Gui, Tao Tian, Le Zhou, Xiao Zheng, Xiaoqing Huang, Xuanjing Zhou, Jie |
| author_facet | Lv, Changze Zhou, Jie Zhao, Wentao Xu, Jingwen Dou, Shihan Huang, Zisu Tian, Muzhao Wang, Xiaohua Liu, Yang Zhou, Pluto Gui, Tao Tian, Le Zhou, Xiao Zheng, Xiaoqing Huang, Xuanjing Zhou, Jie |
| contents | Nowadays, developing reliable DeepResearch-style long-form report generation remains challenging, as training and evaluation lack verifiable reward signals. Accordingly, rubric-based evaluation has become a common practice. However, existing approaches either rely on coarse, pre-defined rubrics that lack sufficient granularity or depend on manually constructed query-specific rubrics that are costly and difficult to scale. In this paper, we propose a pipeline to train preference-grounded query-specific rubric generators tailored for DeepResearch report generation. We first construct a dataset of DeepResearch-style queries annotated with human preferences over paired reports, and train rubric generators via reinforcement learning with a hybrid reward combining preference consistency, format validity, and LLM-based rubric evaluation. We evaluate the resulting rubric generators in two stages. First, on a held-out human-preference test set, the learned rubrics discriminate preferred from rejected reports more effectively than generic, prompted, or SFT-trained rubric alternatives. Second, when used as reward signals to train DeepResearch systems, our rubric generators yield substantial performance gains under both a simple single-agent ReAct framework and a complex multi-agent workflow on the DeepResearch Bench. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03619 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation Lv, Changze Zhou, Jie Zhao, Wentao Xu, Jingwen Dou, Shihan Huang, Zisu Tian, Muzhao Wang, Xiaohua Liu, Yang Zhou, Pluto Gui, Tao Tian, Le Zhou, Xiao Zheng, Xiaoqing Huang, Xuanjing Zhou, Jie Computation and Language Nowadays, developing reliable DeepResearch-style long-form report generation remains challenging, as training and evaluation lack verifiable reward signals. Accordingly, rubric-based evaluation has become a common practice. However, existing approaches either rely on coarse, pre-defined rubrics that lack sufficient granularity or depend on manually constructed query-specific rubrics that are costly and difficult to scale. In this paper, we propose a pipeline to train preference-grounded query-specific rubric generators tailored for DeepResearch report generation. We first construct a dataset of DeepResearch-style queries annotated with human preferences over paired reports, and train rubric generators via reinforcement learning with a hybrid reward combining preference consistency, format validity, and LLM-based rubric evaluation. We evaluate the resulting rubric generators in two stages. First, on a held-out human-preference test set, the learned rubrics discriminate preferred from rejected reports more effectively than generic, prompted, or SFT-trained rubric alternatives. Second, when used as reward signals to train DeepResearch systems, our rubric generators yield substantial performance gains under both a simple single-agent ReAct framework and a complex multi-agent workflow on the DeepResearch Bench. |
| title | Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.03619 |