mlsanityData to 13 Oct 2026

R2-Searcher: Calibrating Retrieval and Reasoning Boundaries for Agentic Search

arXiv 2606.28566 ↗First posted 26 Jun 2026cs.IRcited by 0

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

also in 1 more
Abstract (arXiv)

Recent search agents for multi-hop reasoning often fail by either retrieving incomplete evidence or reasoning over irrelevant portions of the retrieved content, leading to a retrieval-reasoning boundary shift. We propose R2-Searcher, a novel framework that explicitly explores and calibrates the retrieval and reasoning boundaries via fine-grained, query-token-guided evidence modeling and post-retrieval reflection. Specifically, R2-Searcher: (1) constructs fine-grained reasoning contexts by extracting precise facts from retrieved content based on query token semantics (e.g., subjects, actions, temporal markers, and degree modifiers), thereby guiding the attention of search agent; (2) introduces a retrieval reflection mechanism that evaluates and corrects boundary deviations after each retrieval step, guiding the generation of improved queries grounded in the extracted reasoning contexts; and (3) employs an end-to-end reasoning-reflection-guided reinforcement learning algorithm, R2PO, which jointly optimizes both boundaries through a tree-based exploration of reasoning regions and reflections. Our method significantly enhances the quality of both retrieval and reasoning, establishing an iterative loop where retrieval and reasoning mutually enhance each other. Extensive experiments on seven complex multi-hop QA benchmarks demonstrate that R2-Searcher significantly outperforms state-of-the-art agentic search methods in answer accuracy and retrieval-reasoning quality. Ablation studies further confirm the critical role of retrieval-reasoning boundary calibration.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 41 of its 59 references that are in the map. First posted on arXiv 26 Jun 2026.

Builds on

41 of 59 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Interact-RAG: Reason and Interact with the Corpus, Beyond Black-Box Retrieval31 Oct
  2. 2025GraphRAG-R1: Graph Retrieval-Augmented Generation with Process-Constrained Reinforcement Learning31 Jul
  3. 2025Search and Refine During Think: Facilitating Knowledge Refinement for Improved Retrieval-Augmented Reasoning16 May
  4. 2025ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning25 Mar
  5. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning12 Mar
  6. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Models9 Jan

Other references · 35

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,091
  2. 2025Qwen3 Technical Reportcited by 8,510
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,247
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,716
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,412
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,925
  7. 2024Qwen2 Technical Reportcited by 2,102
  8. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,433
  9. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,266
  10. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,229
  11. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,032
  12. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 663
23 more
  1. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexitycited by 442
  2. 2025Zep: A Temporal Knowledge Graph Architecture for Agent Memorycited by 388
  3. 2025MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agentcited by 240
  4. 2025MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agentscited by 230
  5. 2025From RAG to Memory: Non-Parametric Continual Learning for Large Language Modelscited by 226
  6. 2025ZeroSearch: Incentivize the Search Capability of LLMs without Searchingcited by 205
  7. 2023Query2doc: Query Expansion with Large Language Modelscited by 148
  8. 2025G-Memory: Tracing Hierarchical Memory for Multi-Agent Systemscited by 117
  9. 2025Process vs. Outcome Reward: Which is Better for Agentic RAG Reinforcement Learningcited by 68
  10. 2025DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learningcited by 63
  11. 2025Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agentscited by 60
  12. 2025Scaling LLM Multi-turn RL with End-to-end Summarization-based Context Managementcited by 51
  13. 2025Retrieval-Augmented Generation with Hierarchical Knowledgecited by 47
  14. 2025TreeGRPO: Tree-Advantage GRPO for Online RL Post-Training of Diffusion Modelscited by 29
  15. 2025MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learningcited by 27
  16. 2025IterResearch: Rethinking Long-Horizon Agents with Interaction Scalingcited by 23
  17. 2023Embedding in Recommender Systems: A Surveycited by 19
  18. 2025The Markovian Thinker: Architecture-Agnostic Linear Scaling of Reasoningcited by 12
  19. 2025Towards Adaptive Memory-Based Optimization for Enhanced Retrieval-Augmented Generationcited by 9
  20. 2025InfoGain-RAG: Boosting Retrieval-Augmented Generation via Document Information Gain-based Reranking and Filteringcited by 8
  21. 2025KG-Infused RAG: Augmenting Corpus-Based RAG with External Knowledge Graphscited by 4
  22. 2026MemSearch-o1: Empowering Large Language Models with Reasoning-Aligned Memory Growth in Agentic Searchcited by 4
  23. 2026Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimizationcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026D2Plan: Dual-Agent Dynamic Global Planning for Complex Retrieval-Augmented Reasoning
  2. 2026Critic-R: Improving Agentic Search using Instruction-tuned Retrievers with Natural Language Introspective Feedback
  3. 2026Rethinking Reasoning-Intensive Retrieval: Evaluating and Advancing Retrievers in Agentic Search Systems
  4. 2026CoSearch: Joint Training of Reasoning and Document Ranking via Reinforcement Learning for Agentic Search
  5. 2026Agentic-R: Learning to Retrieve for Agentic Search
  6. 2025RECON: Reasoning with Condensation for Efficient Retrieval-Augmented Generation
  7. 2026GRASP: GRanularity-Aware Search Policy for Agentic RAG
  8. 2026PathRouter: Aligning Rewards with Retrieval Quality in Agentic Graph Retrieval-Augmented Generation