FABLE：基於森林的自適應雙路徑大語言模型增強檢索方法——面向多文件推理任務

摘要

長文本大型語言模型的快速發展，重新引發了關於檢索增強生成技術是否仍有必要的討論。然而實證研究顯示，長文本推理仍存在持續性局限，包括中間信息丟失現象、高昂計算成本，以及多文檔推理的可擴展性不足等問題。相比之下，傳統RAG系統雖具效率，但受制於平面塊級檢索機制，這種機制會引入語義噪聲且無法支持結構化的跨文檔綜合分析。我們提出FABLE框架——一種基於森林結構的自適應雙路徑LLM增強檢索系統，將大型語言模型深度整合到知識組織與檢索兩個層面。該框架首先構建具有多粒度語義結構的LLM增強型層次化森林索引，隨後採用雙路徑策略：結合LLM引導的層次化遍歷與結構感知傳播機制進行細粒度證據獲取，並通過顯式預算控制實現自適應的效率平衡。大量實驗表明，FABLE在持續超越現有頂尖RAG方法的同時，能達到與全文本LLM推理相媲美的準確度，且實現高達94%的標記量壓縮。這證實長文本LLM實際強化了而非完全取代對結構化檢索的需求。

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：基於森林的自適應雙路徑大語言模型增強檢索方法——面向多文件推理任務

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

摘要

Support