TAdaRAG：通过动态知识图谱构建实现任务自适应的检索增强生成
TAdaRAG: Task Adaptive Retrieval-Augmented Generation via On-the-Fly Knowledge Graph Construction
AI总结：
针对传统RAG因知识截断丢失信息、非结构化知识引入干扰的问题，提出TAdaRAG框架，通过意图驱动路由、监督微调及强化学习隐式抽取动态构建任务自适应知识图谱，在多基准上优于现有方法，泛化性和实用性强。
摘要：
检索增强生成（RAG）通过检索外部知识提升大型语言模型性能，但外部知识常因输入上下文窗口限制被截断为较小片段，导致信息丢失，引发响应幻觉和推理链断裂。此外，传统RAG检索非结构化知识，引入无关细节阻碍准确推理。为解决这些问题，我们提出TAdaRAG，一种从外部源动态构建任务自适应知识图谱的新型RAG框架。具体而言，我们设计意图驱动的路由机制到特定领域的抽取模板，随后进行监督微调及基于强化学习的隐式抽取机制，确保知识整合简洁、连贯且无冗余。基于三个骨干模型在六个公共基准和一个真实业务基准（NowNewsQA）上的评估显示，TAdaRAG在不同领域和长文本任务中优于现有方法，凸显其强泛化性和实际有效性。
英文摘要：
Retrieval-Augmented Generation (RAG) improves large language models by retrieving external knowledge, often truncated into smaller chunks due to the input context window, which leads to information loss, resulting in response hallucinations and broken reasoning chains. Moreover, traditional RAG retrieves unstructured knowledge, introducing irrelevant details that hinder accurate reasoning. To address these issues, we propose TAdaRAG, a novel RAG framework for on-the-fly task-adaptive knowledge graph construction from external sources. Specifically, we design an intent-driven routing mechanism to a domain-specific extraction template, followed by supervised fine-tuning and a reinforcement learning-based implicit extraction mechanism, ensuring concise, coherent, and non-redundant knowledge integration. Evaluations on six public benchmarks and a real-world business benchmark (NowNewsQA) across three backbone models demonstrate that TAdaRAG outperforms existing methods across diverse domains and long-text tasks, highlighting its strong generalization and practical effectiveness.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
