Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.13059 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909962910826496 |
|---|---|
| author | Yamada, Ikuya Ikeda, Wataru Yoshida, Ko Ye, Mengyu Sugimoto, Hinata Suzuki, Masatoshi Ozaki, Hisanori Suzuki, Jun |
| author_facet | Yamada, Ikuya Ikeda, Wataru Yoshida, Ko Ye, Mengyu Sugimoto, Hinata Suzuki, Masatoshi Ozaki, Hisanori Suzuki, Jun |
| contents | We present an open deep research system for long-form question answering, selected as a winning system in the text-to-text track of the MMU-RAG competition at NeurIPS 2025. The system combines an open-source large language model (LLM) with an open web search API to perform iterative retrieval, reasoning, and synthesis in real-world open-domain settings. To enhance reasoning quality, we apply preference tuning based on LLM-as-a-judge feedback that evaluates multiple aspects, including clarity, insightfulness, and factuality. Our experimental results show that the proposed method consistently improves answer quality across all three aspects. Our source code is publicly available at https://github.com/efficient-deep-research/efficient-deep-research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_13059 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | An Open and Reproducible Deep Research Agent for Long-Form Question Answering Yamada, Ikuya Ikeda, Wataru Yoshida, Ko Ye, Mengyu Sugimoto, Hinata Suzuki, Masatoshi Ozaki, Hisanori Suzuki, Jun Computation and Language We present an open deep research system for long-form question answering, selected as a winning system in the text-to-text track of the MMU-RAG competition at NeurIPS 2025. The system combines an open-source large language model (LLM) with an open web search API to perform iterative retrieval, reasoning, and synthesis in real-world open-domain settings. To enhance reasoning quality, we apply preference tuning based on LLM-as-a-judge feedback that evaluates multiple aspects, including clarity, insightfulness, and factuality. Our experimental results show that the proposed method consistently improves answer quality across all three aspects. Our source code is publicly available at https://github.com/efficient-deep-research/efficient-deep-research. |
| title | An Open and Reproducible Deep Research Agent for Long-Form Question Answering |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.13059 |