Search-o1: Agentic Search-Enhanced Large Reasoning Models
Paper: https://arxiv.org/abs/2501.05366 Code: https://github.com/sunnynexus/Search-o1 Published: 2025-01-09 Authors: Xiaoxi Li, Guanting Dong, Jiajie Jin, Yuyao Zhang, Yujia Zhou, Yutao Zhu, et al. (8 authors)
Overview
Large reasoning models (LRMs) like OpenAI-o1 demonstrate impressive long stepwise reasoning capabilities through large-scale reinforcement learning. However, their extended reasoning processes often suffer from knowledge insufficiency, leading to frequent uncertainties and potential errors.
Search-o1 addresses this limitation with two key components:
1. Agentic RAG mechanism — an agentic search workflow is integrated into the reasoning process, enabling dynamic retrieval of external knowledge when the LRM encounters uncertain knowledge points. 2. Reason-in-Documents module — since retrieved documents are verbose, this separate module deeply analyzes retrieved information before injecting it into the reasoning chain, minimizing noise and preserving a coherent reasoning flow.
Evaluation
Extensive experiments were conducted on:
- Complex reasoning tasks in science, mathematics, and coding
- Six open-domain QA benchmarks
- Synergizing RAG and Reasoning: A Systematic Review (arXiv 2504.15909)
- Agentic Information Retrieval (arXiv 2410.09713)
- AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforcement Learning (arXiv 2509.24193)
Results demonstrate strong performance, enhancing the trustworthiness and applicability of LRMs in complex reasoning tasks and paving the way for more reliable and versatile intelligent systems.
Original Abstract
> Large reasoning models (LRMs) like OpenAI-o1 have demonstrated impressive long stepwise reasoning capabilities through large-scale reinforcement learning. However, their extended reasoning processes often suffer from knowledge insufficiency, leading to frequent uncertainties and potential errors. To address this limitation, we introduce Search-o1, a framework that enhances LRMs with an agentic retrieval-augmented generation (RAG) mechanism and a Reason-in-Documents module for refining retrieved documents. Search-o1 integrates an agentic search workflow into the reasoning process, enabling dynamic retrieval of external knowledge when LRMs encounter uncertain knowledge points. Additionally, due to the verbose nature of retrieved documents, we design a separate Reason-in-Documents module to deeply analyze the retrieved information before injecting it into the reasoning chain, minimizing noise and preserving coherent reasoning flow. Extensive experiments on complex reasoning tasks in science, mathematics, and coding, as well as six open-domain QA benchmarks, demonstrate the strong performance of Search-o1. This approach enhances the trustworthiness and applicability of LRMs in complex reasoning tasks, paving the way for more reliable and versatile intelligent systems. The code is available at https://github.com/sunnynexus/Search-o1.