GRASP: GRanularity-Aware Search Policy for Agentic RAG

• URL: http://arxiv.org/abs/2607.10463v1
• Date: Sat, 11 Jul 2026 19:58:07 GMT
• Title: GRASP: GRanularity-Aware Search Policy for Agentic RAG
• Abstract summary: We introduce GRASP, a reinforcement learning framework for training agents to adaptively coordinate complementary retrieval tools.<n> GRASP provides the agent with semantic search, keyword search, and paragraph-reading actions.<n>We train the policy with a reward that jointly accounts for answer accuracy, grounded reading, complementary search, and turn efficiency.
• Score: 35.781196496632035
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Agentic retrieval-augmented generation (RAG) extends static RAG by allowing language models to iteratively reason, generate search queries, retrieve evidence, and predict answers. However, it remains challenging for models to decide when to retrieve, whether to use lexical matching or semantic similarity, and how to control context granularity to prevent irrelevant tokens from interfering with agent reasoning. In this paper, we introduce GRASP, a reinforcement learning (RL) framework for training agents to adaptively coordinate complementary retrieval tools during multi-step reasoning. GRASP provides the agent with semantic search, keyword search, and paragraph-reading actions, enabling it to retrieve sentence-level evidence and expand further context only when needed. We train the policy with a reward that jointly accounts for answer accuracy, grounded reading, complementary search, and turn efficiency. Experiments on multi-hop reasoning benchmarks show that GRASP improves both retrieval recall and downstream question answering performance compared with single-step retrieval, prompting-based agentic RAG, and RL-based retrieval baselines. Qualitative and ablation analyses show that the learned policy develops interpretable skimming and scanning behavior: it uses semantic search for broad exploration, paragraph reading for local verification, and keyword search for entity-specific evidence. These results suggest that learning to coordinate retrieval signals and context granularity is critical for agent's correct reasoning.

Related papers

• A New Role for Relevance: Guiding Corpus Interaction in Agentic Search [45.65771530564621]
  Relevance is a query-dependent estimate of whether a document or excerpt contains useful evidence.<n>We introduce the Relevance-Aware RipGrep Search Agent (RARG), which turns relevance into an execution prior to interaction.
  arXiv  Detail & Related papers  (2026-07-27T09:56:52Z)
• R$^2$-Searcher: Calibrating Retrieval and Reasoning Boundaries for Agentic Search [59.034878430183305]
  R$2$-Searcher explicitly explores and calibrates the retrieval and reasoning boundaries.<n>We show that R$2$-Searcher significantly outperforms state-of-the-art agentic search methods in answer accuracy and retrieval-reasoning quality.
  arXiv  Detail & Related papers  (2026-06-26T19:44:34Z)
• GrepSeek: Training Search Agents for Direct Corpus Interaction [66.69568141699311]
  GrepSeek is an optimized direct corpus interaction (DCI) search agent that trains a compact search agent to find, filter, and compose evidence from large text corpora.<n>To make DCI practical at scale, we further use a semantics-preserving sharded-parallel execution engine that accelerates shell-based retrieval by up to $7.6times$ while preserving byte-exact equivalence with sequential execution of the shell command.
  arXiv  Detail & Related papers  (2026-05-28T03:37:33Z)
• Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction [127.64173950476702]
  We study direct corpus interaction (DCI), where an agent searches the raw corpus directly with general-purpose terminal tools.<n>This approach requires no offline indexing and adapts naturally to evolving local corpora.<n>Across IR benchmarks and end-to-end agentic search tasks, this simple setup substantially outperforms strong sparse, dense, and reranking baselines.
  arXiv  Detail & Related papers  (2026-05-03T19:13:11Z)
• Cycle-Consistent Search: Question Reconstructability as a Proxy Reward for Search Agent Training [80.20022221643414]
  Cycle-Consistent Search is a gold-supervision-free framework for training search agents.<n>We show that CCS achieves performance comparable to supervised baselines.<n>These results suggest that CCS provides a scalable training paradigm for training search agents in settings where gold supervision is unavailable.
  arXiv  Detail & Related papers  (2026-04-14T17:00:18Z)
• DOTRAG: Retrieval-Time Reasoning Along Paths [36.96341842051057]
  We propose DotRAG, a training-free GraphRAG framework that reformulates retrieval as a reasoning process over paths.<n>Our approach generates query-conditioned constraints that guide graph exploration, prune irrelevant regions, and iteratively discover relational paths without relying on explicit step-by-step reasoning chains.
  arXiv  Detail & Related papers  (2026-04-06T22:38:15Z)
• Do We Still Need GraphRAG? Benchmarking RAG and GraphRAG for Agentic Search Systems [9.226583502132753]
  Retrieval-augmented generation (RAG) and its graph-based extensions (GraphRAG) are effective paradigms for improving large language model (LLM) reasoning.<n>Recent agentic search systems enable dynamic, multi-round retrieval and sequential decision-making during inference.<n>We introduce RAGSearch, a unified benchmark that evaluates dense RAG and representative GraphRAG methods as retrieval infrastructures under agentic search.
  arXiv  Detail & Related papers  (2026-04-01T07:21:32Z)
• Reasoning-Augmented Representations for Multimodal Retrieval [27.4146940988752]
  Universal Multimodal Retrieval (UMR) seeks any-to-any search across text and vision.<n>We argue this brittleness is often data-specified: when images carry "silent" evidence and queries leave key semantics implicit, a single embedding pass must both reason and compress.<n>We propose a data-centric framework that decouples these roles by externalizing reasoning before retrieval.
  arXiv  Detail & Related papers  (2026-02-06T19:01:54Z)
• Chaining the Evidence: Robust Reinforcement Learning for Deep Search Agents with Citation-Aware Rubric Rewards [60.0970117192627]
  Reinforcement learning (RL) has emerged as a critical technique for enhancing LLM-based deep search agents.<n>Existing approaches primarily rely on binary outcome rewards, which fail to capture the comprehensiveness and factuality of agents' reasoning process.<n>We propose textbfCitation-aware RL Rewards (CaRR), a fine-grained reward framework for deep search agents.
  arXiv  Detail & Related papers  (2026-01-09T18:57:53Z)
• ClueAnchor: Clue-Anchored Knowledge Reasoning Exploration and Optimization for Retrieval-Augmented Generation [82.54090885503287]
  Retrieval-Augmented Generation augments Large Language Models with external knowledge to improve factuality.<n>Existing RAG systems fail to extract and integrate the key clues needed to support faithful and interpretable reasoning.<n>We propose ClueAnchor, a novel framework for enhancing RAG via clue-anchored reasoning exploration and optimization.
  arXiv  Detail & Related papers  (2025-05-30T09:18:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.