DeepRAG: Thinking to Retrieve Step by Step for Large Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915332454612992 |
|---|---|
| author | Guan, Xinyan Zeng, Jiali Meng, Fandong Xin, Chunlei Lu, Yaojie Lin, Hongyu Han, Xianpei Sun, Le Zhou, Jie |
| author_facet | Guan, Xinyan Zeng, Jiali Meng, Fandong Xin, Chunlei Lu, Yaojie Lin, Hongyu Han, Xianpei Sun, Le Zhou, Jie |
| contents | Large Language Models (LLMs) have shown remarkable reasoning capabilities, while their practical applications are limited by severe factual hallucinations due to limitations in the timeliness, accuracy, and comprehensiveness of their parametric knowledge. Meanwhile, enhancing retrieval-augmented generation (RAG) with reasoning remains challenging due to ineffective task decomposition and redundant retrieval, which can introduce noise and degrade response quality. In this paper, we propose DeepRAG, a framework that models retrieval-augmented reasoning as a Markov Decision Process (MDP), enabling reasonable and adaptive retrieval. By iteratively decomposing queries, DeepRAG dynamically determines whether to retrieve external knowledge or rely on parametric reasoning at each step. Experiments show that DeepRAG improves retrieval efficiency and boosts answer accuracy by 26.4%, demonstrating its effectiveness in enhancing retrieval-augmented reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01142 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DeepRAG: Thinking to Retrieve Step by Step for Large Language Models Guan, Xinyan Zeng, Jiali Meng, Fandong Xin, Chunlei Lu, Yaojie Lin, Hongyu Han, Xianpei Sun, Le Zhou, Jie Artificial Intelligence Computation and Language Information Retrieval Large Language Models (LLMs) have shown remarkable reasoning capabilities, while their practical applications are limited by severe factual hallucinations due to limitations in the timeliness, accuracy, and comprehensiveness of their parametric knowledge. Meanwhile, enhancing retrieval-augmented generation (RAG) with reasoning remains challenging due to ineffective task decomposition and redundant retrieval, which can introduce noise and degrade response quality. In this paper, we propose DeepRAG, a framework that models retrieval-augmented reasoning as a Markov Decision Process (MDP), enabling reasonable and adaptive retrieval. By iteratively decomposing queries, DeepRAG dynamically determines whether to retrieve external knowledge or rely on parametric reasoning at each step. Experiments show that DeepRAG improves retrieval efficiency and boosts answer accuracy by 26.4%, demonstrating its effectiveness in enhancing retrieval-augmented reasoning. |
| title | DeepRAG: Thinking to Retrieve Step by Step for Large Language Models |
| topic | Artificial Intelligence Computation and Language Information Retrieval |
| url | https://arxiv.org/abs/2502.01142 |