mlsanityData to 14 Oct 2026

AgentFold: Long-Horizon Web Agents with Proactive Context Management

arXiv 2510.24699 ↗First posted 28 Oct 2025cs.CL cs.AI cs.LGcited by 103

On the map: LLM agents and tool use → Agent memory systems

Abstract (arXiv)

LLM-based web agents show immense promise for information seeking, yet their effectiveness on long-horizon tasks is hindered by a fundamental trade-off in context management. Prevailing ReAct-based agents suffer from context saturation as they accumulate noisy, raw histories, while methods that fixedly summarize the full history at each step risk the irreversible loss of critical details. Addressing these, we introduce AgentFold, a novel agent paradigm centered on proactive context management, inspired by the human cognitive process of retrospective consolidation. AgentFold treats its context as a dynamic cognitive workspace to be actively sculpted, rather than a passive log to be filled. At each step, it learns to execute a `folding' operation, which manages its historical trajectory at multiple scales: it can perform granular condensations to preserve vital, fine-grained details, or deep consolidations to abstract away entire multi-step sub-tasks. The results on prominent benchmarks are striking: with simple supervised fine-tuning (without continual pre-training or RL), our AgentFold-30B-A3B agent achieves 36.2% on BrowseComp and 47.3% on BrowseComp-ZH. Notably, this performance not only surpasses or matches open-source models of a dramatically larger scale, such as the DeepSeek-V3.1-671B-A37B, but also surpasses leading proprietary agents like OpenAI's o4-mini.

Key facts. Cited by 103 papers in the map (data through 14 Oct 2026). Placed in Agent memory systems, one of 16 directions in LLM agents and tool use. Builds on 30 of its 39 references that are in the map. First posted on arXiv 28 Oct 2025.

Builds on

30 of 39 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents18 Jun
  2. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory28 Apr
  3. 2025A-MEM: Agentic Memory for LLM Agents17 Feb

Other references · 27

  1. 2025Qwen3 Technical Reportcited by 8,820
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  5. 2023GAIA: a benchmark for General AI Assistantscited by 1,128
  6. 2025BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agentscited by 647
  7. 2025GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Modelscited by 467
  8. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 335
  9. 2025MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agentcited by 246
  10. 2025WebSailor: Navigating Super-human Reasoning for Web Agentcited by 229
  11. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  12. 2025WebDancer: Towards Autonomous Information Seeking Agencycited by 188
15 more
  1. 2025Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RLcited by 150
  2. 2025BrowseComp-ZH: Benchmarking Web Browsing Ability of Large Language Models in Chinesecited by 145
  3. 2025A Survey of Context Engineering for Large Language Modelscited by 126
  4. 2025WebShaper: Agentically Data Synthesizing via Information-Seeking Formalizationcited by 121
  5. 2025WebExplorer: Explore and Evolve for Training Long-Horizon Web Agentscited by 92
  6. 2025MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Modelscited by 73
  7. 2025WideSearch: Benchmarking Agentic Broad Info-Seekingcited by 69
  8. 2025WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agentscited by 68
  9. 2025WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learningcited by 61
  10. 2025DeepDive: Advancing Deep Search Agents with Knowledge Graphs and Multi-Turn RLcited by 55
  11. 2025MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systemscited by 50
  12. 2025SciMaster: Towards General-Purpose Scientific AI Agents, Part I. X-Master as Foundation: Can We Lead on Humanity's Last Exam?cited by 47
  13. 2024Memory3: Language Modeling with Explicit Memorycited by 40
  14. 2025DeepDiver: Adaptive Search Intensity Scaling via Open-Web Reinforcement Learningcited by 23
  15. 2025BrowseMaster: Towards Scalable Web Browsing via Tool-Augmented Programmatic Agent Paircited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 96.

103 papers.

Sort:

By area:

  1. 2025Memory in the Age of AI Agentscited by 303
  2. 2025Recursive Language Modelscited by 104
  3. 2025Budget-Aware Tool Use Enables Effective Agent Scalingcited by 44
  4. 2026SWE-Pruner: Self-Adaptive Context Pruning for Coding Agentscited by 44
  5. 2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agentscited by 44
  6. 2025Context as a Tool: Context Management for Long-Horizon SWE-Agentscited by 43
  7. 2026OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Datacited by 39
  8. 2026Benchmark Test-Time Scaling of General LLM Agentscited by 26
  9. 2026From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanismscited by 21
  10. 2026Mem-T: Densifying Rewards for Long-Horizon Memory Agentscited by 18
  11. 2026Toward Efficient Agents: Memory, Tool learning, and Planningcited by 16
  12. 2026JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolutioncited by 16
38 more
  1. 2026AtomMem : Learnable Dynamic Agentic Memory with Atomic Memory Operationcited by 15
  2. 2026A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compressioncited by 15
  3. 2026SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agentscited by 15
  4. 2026ContextBudget: Budget-Aware Context Management for Long-Horizon Search Agentscited by 14
  5. 2026MemoBrain: Executive Memory as an Agentic Brain for Reasoningcited by 13
  6. 2025AI Meets Brain: Memory Systems from Cognitive Neuroscience to Autonomous Agentscited by 12
  7. 2026Memex(RL): Scaling Long-Horizon LLM Agents via Indexed Experience Memorycited by 12
  8. 2026Coding Agents are Effective Long-Context Processorscited by 12
  9. 2026VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactionscited by 12
  10. 2026ToolCUA: Towards Optimal GUI-Tool Path Orchestration for Computer Use Agentscited by 11
  11. 2026Self-Compacting Language Model Agentscited by 11
  12. 2026LongSeeker: Elastic Context Orchestration for Long-Horizon Search Agentscited by 10
  13. 2026SAM: State-Adaptive Memory for Long-Horizon Reasoning Agentcited by 10
  14. 2026Understanding the Challenges in Iterative Generative Optimization with LLMscited by 8
  15. 2026Scaling Coding Agents via Atomic Skillscited by 8
  16. 2026OpenSeeker-v2: Pushing the Limits of Search Agents with Informative and High-Difficulty Trajectoriescited by 8
  17. 2026Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Applicationcited by 8
  18. 2026Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflictscited by 7
  19. 2026ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agentscited by 7
  20. 2025FoldAct: Efficient and Stable Context Folding for Long-Horizon Search Agentscited by 6
  21. 2026Gated Differentiable Working Memory for Long-Context Language Modelingcited by 6
  22. 2026DLLM Agent: See Farther, Run Fastercited by 6
  23. 2026When the Specification Emerges: Benchmarking Faithfulness Loss in Long-Horizon Coding Agentscited by 6
  24. 2026SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Contextcited by 6
  25. 2026Argus: Evidence Assembly for Scalable Deep Research Agentscited by 6
  26. 2025PRInTS: Reward Modeling for Long-Horizon Information Seekingcited by 5
  27. 2026SwiftMem: Fast Agentic Memory via Query-aware Indexingcited by 5
  28. 2026VimRAG: Navigating Massive Visual Context in Retrieval-Augmented Generation via Multimodal Memory Graphcited by 5
  29. 2026CoMem: Context Management with A Decoupled Long-Context Modelcited by 5
  30. 2026Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanismcited by 5
  31. 2026SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harnesscited by 5
  32. 2026HyMem: Hybrid Memory Architecture with Dynamic Retrieval Schedulingcited by 4
  33. 2026LightThinker++: From Reasoning Compression to Memory Managementcited by 4
  34. 2026Time is Not a Label: Continuous Phase Rotation for Temporal Knowledge Graphs and Agentic Memorycited by 4
  35. 2026Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memorycited by 4
  36. 2026CoACT: Action-Preserving Observation Compression for Coding Agentscited by 4
  37. 2026Context as an Environment: Programmatic Context Management for Long-Horizon Agentscited by 4
  38. 2026Yunque DeepResearch Technical Reportcited by 3

50 most cited of 103 All 103 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026ACM: Agentic Context Management for Long Horizon Tasks
  2. 2026WebChallenger: A Reliable and Efficient Generalist Web Agent
  3. 2026LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues
  4. 2026Beyond Compaction: Structured Context Eviction for Long-Horizon Agents
  5. 2026AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agents
  6. 2025Lost in the Maze: Overcoming Context Limitations in Long-Horizon Agentic Search
  7. 2025ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
  8. 2025WebCoach: Self-Evolving Web Agents with Cross-Session Memory Guidance