DIVER: A Multi-Stage Approach for Reasoning-intensive Information Retrieval
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911561261514752 |
|---|---|
| author | Sun, Duolin Long, Meixiu Yang, Dan Wang, Junjie Luo, Yecheng Shen, Yue Wang, Jian Zhou, Hualei Guo, Chunxiao Wei, Peng Wang, Jiahai Gu, Jinjie |
| author_facet | Sun, Duolin Long, Meixiu Yang, Dan Wang, Junjie Luo, Yecheng Shen, Yue Wang, Jian Zhou, Hualei Guo, Chunxiao Wei, Peng Wang, Jiahai Gu, Jinjie |
| contents | Retrieval-augmented generation has achieved strong performance on knowledge-intensive tasks where query-document relevance can be identified through direct lexical or semantic matches. However, many real-world queries involve abstract reasoning, analogical thinking, or multi-step inference, which existing retrievers often struggle to capture. To address this challenge, we present DIVER, a retrieval pipeline designed for reasoning-intensive information retrieval. It consists of four components. The document preprocessing stage enhances readability and preserves content by cleaning noisy texts and segmenting long documents. The query expansion stage leverages large language models to iteratively refine user queries with explicit reasoning and evidence from retrieved documents. The retrieval stage employs a model fine-tuned on synthetic data spanning medical and mathematical domains, along with hard negatives, enabling effective handling of reasoning-intensive queries. Finally, the reranking stage combines pointwise and listwise strategies to produce both fine-grained and globally consistent rankings. On the BRIGHT benchmark, DIVER achieves state-of-the-art nDCG@10 scores of 46.8 overall and 31.9 on original queries, consistently outperforming competitive reasoning-aware models. These results demonstrate the effectiveness of reasoning-aware retrieval strategies in complex real-world tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_07995 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DIVER: A Multi-Stage Approach for Reasoning-intensive Information Retrieval Sun, Duolin Long, Meixiu Yang, Dan Wang, Junjie Luo, Yecheng Shen, Yue Wang, Jian Zhou, Hualei Guo, Chunxiao Wei, Peng Wang, Jiahai Gu, Jinjie Information Retrieval Artificial Intelligence Retrieval-augmented generation has achieved strong performance on knowledge-intensive tasks where query-document relevance can be identified through direct lexical or semantic matches. However, many real-world queries involve abstract reasoning, analogical thinking, or multi-step inference, which existing retrievers often struggle to capture. To address this challenge, we present DIVER, a retrieval pipeline designed for reasoning-intensive information retrieval. It consists of four components. The document preprocessing stage enhances readability and preserves content by cleaning noisy texts and segmenting long documents. The query expansion stage leverages large language models to iteratively refine user queries with explicit reasoning and evidence from retrieved documents. The retrieval stage employs a model fine-tuned on synthetic data spanning medical and mathematical domains, along with hard negatives, enabling effective handling of reasoning-intensive queries. Finally, the reranking stage combines pointwise and listwise strategies to produce both fine-grained and globally consistent rankings. On the BRIGHT benchmark, DIVER achieves state-of-the-art nDCG@10 scores of 46.8 overall and 31.9 on original queries, consistently outperforming competitive reasoning-aware models. These results demonstrate the effectiveness of reasoning-aware retrieval strategies in complex real-world tasks. |
| title | DIVER: A Multi-Stage Approach for Reasoning-intensive Information Retrieval |
| topic | Information Retrieval Artificial Intelligence |
| url | https://arxiv.org/abs/2508.07995 |