{"ID":2839816,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.14096","arxiv_id":"2511.14096","title":"NeuroPath: Neurobiology-Inspired Path Tracking and Reflection for Semantically Coherent Retrieval","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.","short_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 capt...","url_abs":"https://arxiv.org/abs/2511.14096","url_pdf":"https://arxiv.org/pdf/2511.14096v1","authors":"[\"Junchen Li\",\"Rongzheng Wang\",\"Yihong Huang\",\"Qizhi Chen\",\"Jiasheng Zhang\",\"Shuang Liang\"]","published":"2025-11-18T03:28:23Z","proceeding":"cs.IR","tasks":"[\"cs.IR\",\"cs.AI\"]","methods":"[\"RAG\",\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":606909,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2839816,"paper_url":"https://arxiv.org/abs/2511.14096","paper_title":"NeuroPath: Neurobiology-Inspired Path Tracking and Reflection for Semantically Coherent Retrieval","repo_url":"https://github.com/KennyCaty/NeuroPath","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}