让大模型边推理边生成搜索向量，提升检索效果
Search-R3: Unifying Reasoning and Embedding in Large Language Models
- 大模型推理时直接输出搜索向量，融合思维链与语义表示
- 在多个基准上显著超越现有方法，最高提升18.7%的检索准确率
- 适合需要复杂推理与精准检索的智能问答系统
尽管大型语言模型具备强大的自然语言理解能力，但在检索任务中仍未被充分使用。我们提出Search-R3，一种新框架，通过将大模型适配为在推理过程中直接生成搜索嵌入向量，解决这一局限。该方法利用大模型的思维链能力，通过逐步进行复杂的语义分析，生成更有效的嵌入表示。具体实现包含三个互补机制：(1) 监督学习阶段提升模型生成高质量嵌入的能力；(2) 强化学习（RL）方法同步优化嵌入生成与推理过程；(3) 专用强化学习环境，高效处理动态嵌入表示，无需每次训练迭代重新编码整个语料库。我们在多种基准上进行广泛评估，结果表明Search-R3显著优于先前方法，通过统一推理与嵌入生成流程，在复杂知识密集型任务中实现显著性能提升。该后训练方法代表了处理需结合深度推理与有效信息检索任务的重要进展。
原文摘要 · Abstract (English)
Despite their remarkable natural language understanding capabilities, Large Language Models (LLMs) have been underutilized for retrieval tasks. We present Search-R3, a novel framework that addresses this limitation by adapting LLMs to generate search embeddings as a direct output of their reasoning process. Our approach exploits LLMs' chain-of-thought capabilities, allowing them to produce more effective embeddings by reasoning step-by-step through complex semantic analyses. We implement this through three complementary mechanisms. (1) a supervised learning stage enables the model's ability to produce quality embeddings, (2) a reinforcement learning (RL) methodology that optimizes embedding generation alongside reasoning, and (3) a specialized RL environment that efficiently handles evolving embedding representations without requiring complete corpus re-encoding at each training iteration. Our extensive evaluations on diverse benchmarks demonstrate that Search-R3 significantly outperforms prior methods by unifying the reasoning and embedding generation processes. This integrated post-training approach represents a substantial advancement in handling complex knowledge-intensive tasks that require both sophisticated reasoning and effective information retrieval. Project page: https://github.com/ytgui/Search-R3
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。