FABLE：基于森林的自适应双路径大语言模型增强检索在多文档推理中的应用

摘要

长上下文大语言模型（LLMs）的快速发展，重新引发了关于检索增强生成（RAG）是否仍有必要的讨论。然而实证研究表明，长上下文推理仍存在固有局限，包括中间信息丢失现象、高昂的计算成本以及多文档推理的可扩展性差等问题。相比之下，传统RAG系统虽然高效，但受限于平面分块检索机制，这种机制会引入语义噪声且无法支持结构化的跨文档综合。我们提出FABLE框架——一种基于森林的自适应双路径LLM增强检索系统，将LLMs深度融合到知识组织与检索两个层面。该框架首先构建具有多粒度语义结构的LLM增强型层次化森林索引，随后采用双路径检索策略：结合LLM引导的层次化遍历与结构感知传播机制实现细粒度证据获取，并通过显式预算控制实现自适应效率权衡。大量实验表明，FABLE不仅持续超越现有最优RAG方法，更在减少高达94%令牌消耗的同时达到与全上下文LLM推理相当的精度。这证明长上下文LLMs实际上强化而非完全取代了对结构化检索的需求。

English

The rapid expansion of long-context Large Language Models (LLMs) has reignited debate on whether Retrieval-Augmented Generation (RAG) remains necessary. However, empirical evidence reveals persistent limitations of long-context inference, including the lost-in-the-middle phenomenon, high computational cost, and poor scalability for multi-document reasoning. Conversely, traditional RAG systems, while efficient, are constrained by flat chunk-level retrieval that introduces semantic noise and fails to support structured cross-document synthesis. We present FABLE, a Forest-based Adaptive Bi-path LLM-Enhanced retrieval framework that integrates LLMs into both knowledge organization and retrieval. FABLE constructs LLM-enhanced hierarchical forest indexes with multi-granularity semantic structures, then employs a bi-path strategy combining LLM-guided hierarchical traversal with structure-aware propagation for fine-grained evidence acquisition, with explicit budget control for adaptive efficiency trade-offs. Extensive experiments demonstrate that FABLE consistently outperforms SOTA RAG methods and achieves comparable accuracy to full-context LLM inference with up to 94\% token reduction, showing that long-context LLMs amplify rather than fully replace the need for structured retrieval.

FABLE：基于森林的自适应双路径大语言模型增强检索在多文档推理中的应用

FABLE: Forest-Based Adaptive Bi-Path LLM-Enhanced Retrieval for Multi-Document Reasoning

摘要

Support