NeuroPath: Neurobiology-Inspired Path Tracking and Reflection for Semantically Coherent Retrieval

TL;DR

NeuroPath introduces a neurobiology-inspired semantic path tracking framework for retrieval-augmented generation, significantly improving multi-hop question answering by enhancing semantic coherence and reducing noise in knowledge graph-based retrieval.

Contribution

The paper presents NeuroPath, a novel LLM-driven framework that improves multi-hop retrieval by dynamic path tracking and post-retrieval refinement, inspired by neurobiological navigation mechanisms.

Findings

Achieves 16.3% higher recall@2 on multi-hop QA datasets.

Reduces token consumption by 22.8% compared to existing methods.

Demonstrates robustness across multiple smaller LLMs and task complexities.

Abstract

Retrieval-augmented generation (RAG) greatly enhances large language models (LLMs) performance in knowledge-intensive tasks. However, naive RAG methods struggle with multi-hop question answering due to their limited capacity to capture complex dependencies across documents. Recent studies employ graph-based RAG to capture document connections. However, these approaches often result in a loss of semantic coherence and introduce irrelevant noise during node matching and subgraph construction. To address these limitations, we propose NeuroPath, an LLM-driven semantic path tracking RAG framework inspired by the path navigational planning of place cells in neurobiology. It consists of two steps: Dynamic Path Tracking and Post-retrieval Completion. Dynamic Path Tracking performs goal-directed semantic path tracking and pruning over the constructed knowledge graph (KG), improving noise reduction and semantic coherence. Post-retrieval Completion further reinforces these benefits by conducting second-stage retrieval using intermediate reasoning and the original query to refine the query goal and complete missing information in the reasoning path. NeuroPath surpasses current state-of-the-art baselines on three multi-hop QA datasets, achieving average improvements of 16.3% on recall@2 and 13.5% on recall@5 over advanced graph-based RAG methods. Moreover, compared to existing iter-based RAG methods, NeuroPath achieves higher accuracy and reduces token consumption by 22.8%. Finally, we demonstrate the robustness of NeuroPath across four smaller LLMs (Llama3.1, GLM4, Mistral0.3, and Gemma3), and further validate its scalability across tasks of varying complexity. Code is available at https://github.com/KennyCaty/NeuroPath.