ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization

• URL: http://arxiv.org/abs/2509.13313v2
• Date: Wed, 15 Oct 2025 15:51:13 GMT
• Title: ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
• Abstract summary: Large Language Model (LLM)-based web agents demonstrate strong performance on knowledge-intensive tasks.<n>Complex queries involving multiple entities, intertwined relationships, and high uncertainty demand extensive search cycles.<n>We introduce ReSum, a novel paradigm that enables indefinite exploration through periodic context summarization.
• Score: 86.57743637722596
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Large Language Model (LLM)-based web agents demonstrate strong performance on knowledge-intensive tasks but are hindered by context window limitations in paradigms like ReAct. Complex queries involving multiple entities, intertwined relationships, and high uncertainty demand extensive search cycles that rapidly exhaust context budgets before reaching solutions. To overcome this challenge, we introduce ReSum, a novel paradigm that enables indefinite exploration through periodic context summarization. ReSum converts growing interaction histories into compact reasoning states, maintaining awareness of prior discoveries while bypassing context constraints. For paradigm adaptation, we propose ReSum-GRPO, integrating GRPO with segmented trajectory training and advantage broadcasting to familiarize agents with summary-conditioned reasoning. Extensive experiments on web agents across three benchmarks demonstrate that ReSum delivers an average absolute improvement of 4.5% over ReAct, with further gains of 8.2% following ReSum-GRPO training. Notably, with only 1K training samples, our WebResummer-30B (a ReSum-GRPO-trained version of WebSailor-30B) achieves 33.3% Pass@1 on BrowseComp-zh and 18.3% on BrowseComp-en, surpassing most open-source web agents.

Related papers

• Verbal-R3: Verbal Reranker as the Missing Bridge between Retrieval and Reasoning [48.15732775780642]
  This paper proposes to bridge retrieval results and the Large Language Model's reasoning ability through Verbal s.<n>We introduce Verbal-R3, a novel agentic RAG framework that consists of a Generator and a Verbal Reranker.<n> Verbal-R3 achieves state-of-the-art performance on complex Question Answering benchmarks, validating the effectiveness of the proposed framework.
  arXiv  Detail & Related papers  (2026-05-02T11:43:23Z)
• CLEAR: Context Augmentation from Contrastive Learning of Experience via Agentic Reflection [17.486111179784757]
  We propose a generative context augmentation framework using Contrastive Learning of Experience via Agentic Reflection (CLEAR)<n>CLEAR first employs a reflection agent to perform contrastive analysis over past execution trajectories and summarize useful context for each observed task.<n>We further optimize CAM using reinforcement learning, where the reward signal is obtained by running the task execution agent.
  arXiv  Detail & Related papers  (2026-04-08T18:26:59Z)
• RetroAgent: From Solving to Evolving via Retrospective Dual Intrinsic Feedback [54.39884046754265]
  RetroAgent is an online RL framework that enables agents to master complex interactive environments.<n>Our experiments show that RetroAgent achieves state-of-the-art (SOTA) performance.
  arXiv  Detail & Related papers  (2026-03-09T16:23:33Z)
• ReAG: Reasoning-Augmented Generation for Knowledge-based Visual Question Answering [54.72902502486611]
  ReAG is a Reasoning-Augmented Multimodal RAG approach that combines coarse- and fine-grained retrieval with a critic model that filters irrelevant passages.<n>ReAG significantly outperforms prior methods, improving answer accuracy and providing interpretable reasoning grounded in retrieved evidence.
  arXiv  Detail & Related papers  (2025-11-27T19:01:02Z)
• IterResearch: Rethinking Long-Horizon Agents via Markovian State Reconstruction [107.49922328855025]
  IterResearch is a novel iterative deep-research paradigm that reformulates long-horizon research as a Markov Decision Process.<n>It achieves substantial improvements over existing open-source agents with average +14.5pp across six benchmarks.<n>It serves as an effective prompting strategy, improving frontier models by up to 19.2pp over ReAct on long-horizon tasks.
  arXiv  Detail & Related papers  (2025-11-10T17:30:08Z)
• Empowering RepoQA-Agent based on Reinforcement Learning Driven by Monte-carlo Tree Search [70.63903518295785]
  We introduce RepoSearch-R1, a novel agentic reinforcement learning framework driven by Monte-carlo Tree Search.<n>Based on RepoSearch-R1, we construct a RepoQA-Agent specifically designed for repository question-answering tasks.
  arXiv  Detail & Related papers  (2025-10-30T09:10:36Z)
• Beyond Turn Limits: Training Deep Search Agents with Dynamic Context Window [88.85901839023803]
  DeepMiner is a novel framework that elicits such abilities by introducing high-difficulty training tasks and dynamic context window.<n>We develop DeepMiner-32B, which achieves substantial performance improvements across multiple search agent benchmarks.
  arXiv  Detail & Related papers  (2025-10-09T14:31:39Z)
• Reasoning-enhanced Query Understanding through Decomposition and Interpretation [130.19204432111277]
  ReDI is a Reasoning-enhanced approach for query understanding through Decomposition and Interpretation.<n>We compiled a large-scale dataset of real-world complex queries from a major search engine.<n>Experiments on BRIGHT and BEIR demonstrate that ReDI consistently surpasses strong baselines in both sparse and dense retrieval paradigms.
  arXiv  Detail & Related papers  (2025-09-08T10:58:42Z)
• Towards Open-World Retrieval-Augmented Generation on Knowledge Graph: A Multi-Agent Collaboration Framework [21.896955284099334]
  Large Language Models (LLMs) have demonstrated strong capabilities in language understanding and reasoning.<n>Retrieval-Augmented Generation (RAG) addresses this limitation by incorporating external knowledge sources.<n>We propose AnchorRAG, a novel multi-agent collaboration framework for open-world RAG without the predefined anchor entities.
  arXiv  Detail & Related papers  (2025-09-01T08:26:12Z)
• REFRAG: Rethinking RAG based Decoding [67.4862300145604]
  REFRAG is an efficient decoding framework that compresses, senses, and expands to improve latency in RAG applications.<n>We provide rigorous validation of REFRAG across diverse long-context tasks, including RAG, multi-turn conversations, and long document summarization.
  arXiv  Detail & Related papers  (2025-09-01T03:31:44Z)
• Training Long-Context, Multi-Turn Software Engineering Agents with Reinforcement Learning [29.605396813225386]
  We show how reinforcement learning can be used to train agents for multi-turn interactive tasks.<n>Our methodology offers a practical approach for training capable agents for multi-turn interactive tasks using open-weight models.
  arXiv  Detail & Related papers  (2025-08-05T14:30:47Z)
• Contextual Experience Replay for Self-Improvement of Language Agents [47.51006612841945]
  We propose Contextual Experience Replay (CER) to enable efficient self-improvement for language agents.<n>CER accumulates and synthesizes past experiences into a dynamic memory buffer.<n>We evaluate CER on the challenging WebArena and VisualWebArena benchmarks.
  arXiv  Detail & Related papers  (2025-06-07T07:47:35Z)
• UniversalRAG: Retrieval-Augmented Generation over Corpora of Diverse Modalities and Granularities [53.76854299076118]
  UniversalRAG is a novel RAG framework designed to retrieve and integrate knowledge from heterogeneous sources with diverse modalities and granularities.<n>We propose a modality-aware routing mechanism that dynamically identifies the most appropriate modality-specific corpus and performs targeted retrieval within it.<n>We validate UniversalRAG on 8 benchmarks spanning multiple modalities, showing its superiority over various modality-specific and unified baselines.
  arXiv  Detail & Related papers  (2025-04-29T13:18:58Z)
• A New Query Expansion Approach via Agent-Mediated Dialogic Inquiry [10.76224743599566]
  We propose AMD: a new Agent-Mediated Dialogic Framework that engages in a dialogic inquiry involving three specialized roles.<n>By leveraging a multi-agent process, AMD effectively crafts richer query representations through inquiry and feedback refinement.
  arXiv  Detail & Related papers  (2025-02-12T16:39:06Z)
• Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning [27.867369806400834]
  We propose RAGPT, a novel Retrieval-AuGmented dynamic Prompt Tuning framework.<n>RAGPT comprises three modules: (I) the multi-channel retriever, (II) the missing modality generator, and (III) the context-aware prompter.<n>Experiments conducted on three real-world datasets show that RAGPT consistently outperforms all competitive baselines in handling incomplete modality problems.
  arXiv  Detail & Related papers  (2025-01-02T07:39:48Z)
• RQ-RAG: Learning to Refine Queries for Retrieval Augmented Generation [42.82192656794179]
  Large Language Models (LLMs) exhibit remarkable capabilities but are prone to generating inaccurate or hallucinatory responses. This limitation stems from their reliance on vast pretraining datasets, making them susceptible to errors in unseen scenarios. Retrieval-Augmented Generation (RAG) addresses this by incorporating external, relevant documents into the response generation process.
  arXiv  Detail & Related papers  (2024-03-31T08:58:54Z)
• Enriching Relation Extraction with OpenIE [70.52564277675056]
  Relation extraction (RE) is a sub-discipline of information extraction (IE) In this work, we explore how recent approaches for open information extraction (OpenIE) may help to improve the task of RE. Our experiments over two annotated corpora, KnowledgeNet and FewRel, demonstrate the improved accuracy of our enriched models.
  arXiv  Detail & Related papers  (2022-12-19T11:26:23Z)

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.