mlsanityData to 14 Oct 2026

DeepSieve: Information Sieving via LLM-as-a-Knowledge-Router

arXiv 2507.22050 ↗First posted 29 Jul 2025cs.CLcited by 13

On the map: LLM reasoning → Knowledge- and tool-augmented reasoning

also in 1 more
Abstract (arXiv)

Large Language Models (LLMs) excel at many reasoning tasks but struggle with knowledge-intensive queries due to their inability to dynamically access up-to-date or domain-specific information. Retrieval-Augmented Generation (RAG) has emerged as a promising solution, enabling LLMs to ground their responses in external sources. However, existing RAG methods lack fine-grained control over both the query and source sides, often resulting in noisy retrieval and shallow reasoning. In this work, we introduce DeepSieve, an agentic RAG framework that incorporates information sieving via LLM-as-a-knowledge-router. DeepSieve decomposes complex queries into structured sub-questions and recursively routes each to the most suitable knowledge source, filtering irrelevant information through a multi-stage distillation process. Our design emphasizes modularity, transparency, and adaptability, leveraging recent advances in agentic system design. Experiments on multi-hop QA tasks across heterogeneous sources demonstrate improved reasoning depth, retrieval precision, and interpretability over conventional RAG approaches. Our codes are available at https://github.com/MinghoKwok/DeepSieve.

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 38 of its 59 references that are in the map. First posted on arXiv 29 Jul 2025.

Builds on

38 of 59 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models1 Jun
  2. 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions20 Dec

Other references · 36

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  5. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  6. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  7. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  8. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,453
  9. 2020What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Examscited by 1,356
  10. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,282
  11. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,210
  12. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,063
24 more
  1. 2017Reading Wikipedia to Answer Open-Domain Questionscited by 777
  2. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 558
  3. 2024RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrievalcited by 532
  4. 2021ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interactioncited by 476
  5. 2022Selection-Inference: Exploiting Large Language Models for Interpretable Logical Reasoningcited by 334
  6. 2024G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answeringcited by 223
  7. 2021When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Datasetcited by 201
  8. 2023ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Modelscited by 184
  9. 2021MultiModalQA: Complex Question Answering over Text, Tables and Imagescited by 179
  10. 2023ADaPT: As-Needed Decomposition and Planning with Language Modelscited by 177
  11. 2024Visual Agents as Fast and Slow Thinkerscited by 53
  12. 2021Decomposing Complex Questions Makes Multi-Hop QA Easier and More Interpretablecited by 35
  13. 2025OmniRouter: Budget and Performance Controllable Multi-LLM Routingcited by 35
  14. 2024Probing-RAG: Self-Probing to Guide Language Models in Selective Document Retrievalcited by 29
  15. 2025AutoSchemaKG: Autonomous Knowledge Graph Construction through Dynamic Schema Induction from Web-Scale Corporacited by 23
  16. 2024Towards Knowledge Checking in Retrieval-augmented Generation: A Representation Perspectivecited by 20
  17. 2024InfuserKI: Enhancing Large Language Models with Knowledge Graphs via Infuser-Guided Knowledge Integrationcited by 19
  18. 2025SRAG: Structured Retrieval-Augmented Generation for Multi-Entity Question Answering over Wikipedia Graphcited by 11
  19. 2025Challenges and Innovations in LLM-Powered Fake News Detection: A Synthesis of Approaches and Future Directionscited by 9
  20. 2023Retrieval-Augmented Chain-of-Thought in Semi-structured Domainscited by 4
  21. 2025LiteCUA: Computer as MCP Server for Computer-Use Agent on AIOScited by 4
  22. 2025ValueSim: Generating Backstories to Model Individual Value Systemscited by 4
  23. 2025Cardiverse: Harnessing LLMs for Novel Card Game Prototypingcited by 3
  24. 2025Towards Efficient Key-Value Cache Management for Prefix Prefilling in LLM Inferencecited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 10.

13 papers.

Sort:

By area:

  1. 2025Cache Mechanism for Agent RAG Systemscited by 5
  2. 2026Skill-RAG: Failure-State-Aware Retrieval Augmentation via Hidden-State Probing and Skill Routingcited by 3
  3. 2025Attention Basin: Why Contextual Position Matters in Large Language Modelscited by 2
  4. 2026VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Modelscited by 2
  5. 2026RASER: Recoverability-Aware Selective Escalation Router for Multi-Hop Question Answeringcited by 2
  6. 2026Individual Turing Test: A Case Study of LLM-based Simulation Using Longitudinal Personal Datacited by 1
  7. 2026Conformal Path Reasoning: Trustworthy Knowledge Graph Question Answering via Path-Level Calibrationcited by 1
  8. 2026Trust or Abstain? A Self-Aware RAG Approachcited by 1
  9. 2025Improving LLM Reasoning via Dependency-Aware Query Decomposition and Logic-Parallel Content Expansioncited by 0
  10. 2026Orion-RAG: Path-Aligned Hybrid Retrieval for Graphless Datacited by 0
  11. 2026RealRoute: Dynamic Query Routing System via Retrieve-then-Verify Paradigmcited by 0
  12. 2026OpenPM: Auditable Point-in-Time Evaluation for LLM Portfolio-Management Agentscited by 0
1 more
  1. 2026QUMem: Personalized Memory for Query-Conditioned User-State Inference in LLM Agentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025RAGCap-Bench: Benchmarking Capabilities of LLMs in Agentic Retrieval Augmented Generation Systems
  2. 2025Towards Agentic RAG with Deep Reasoning: A Survey of RAG-Reasoning Systems in LLMs
  3. 2025MA-RAG: Multi-Agent Retrieval-Augmented Generation via Collaborative Chain-of-Thought Reasoning
  4. 2025SIRAG: Towards Stable and Interpretable RAG with A Process-Supervised Multi-Agent Framework
  5. 2025Agent-UniRAG: A Trainable Open-Source LLM Agent Framework for Unified Retrieval-Augmented Generation Systems
  6. 2026Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval
  7. 2025QAgent: A modular Search Agent with Interactive Query Understanding
  8. 2026Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model