IGMiRAG: Intuition-Guided Retrieval-Augmented Generation with Adaptive Mining of In-Depth Memory

2026-02-07

AI Summary

This work addresses the high retrieval cost and fragmentation in existing Retrieval-Augmented Generation (RAG) methods, which stem from misaligned memory organization. Inspired by human intuitive reasoning, the authors propose an intuition-guided RAG framework that aligns multi-granularity knowledge through a hierarchical heterogeneous hypergraph. A query parser dynamically adjusts the depth and window of memory exploration, while a dual-focus retrieval mechanism activates immediate memory anchors. Bidirectional diffusion algorithms further simulate deductive reasoning paths. This approach is the first to integrate human-like reasoning mechanisms into memory structure and resource allocation, enabling task-complexity-adaptive efficient retrieval. It outperforms the strongest baseline by 4.8% in EM and 5.0% in F1, while substantially reducing token consumption—averaging 6.3k tokens and as low as 3.0k.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Retrieval-augmented generation (RAG) equips large language models (LLMs) with reliable knowledge memory. To strengthen cross-text associations, recent research integrates graphs and hypergraphs into RAG to capture pairwise and multi-entity relations as structured links. However, their misaligned memory organization necessitates costly, disjointed retrieval. To address these limitations, we propose IGMiRAG, a framework inspired by human intuition-guided reasoning. It constructs a hierarchical heterogeneous hypergraph to align multi-granular knowledge, incorporating deductive pathways to simulate realistic memory structures. During querying, IGMiRAG distills intuitive strategies via a question parser to control mining depth and memory window, and activates instantaneous memories as anchors using dual-focus retrieval. Mirroring human intuition, the framework guides retrieval resource allocation dynamically. Furthermore, we design a bidirectional diffusion algorithm that navigates deductive paths to mine in-depth memories, emulating human reasoning processes. Extensive evaluations indicate IGMiRAG outperforms the state-of-the-art baseline by 4.8% EM and 5.0% F1 overall, with token costs adapting to task complexity (average 6.3k+, minimum 3.0k+). This work presents a cost-effective RAG paradigm that improves both efficiency and effectiveness.

Problem

• Retrieval-Augmented Generation
• memory organization
• cross-text associations
• structured knowledge
• inefficient retrieval

Innovation

• Intuition-Guided Reasoning
• Hierarchical Heterogeneous Hypergraph
• Dual-Focus Retrieval
• Bidirectional Diffusion Algorithm
• Adaptive Memory Mining