mlsanityData to 14 Oct 2026

WebSailor: Navigating Super-human Reasoning for Web Agent

arXiv 2507.02592 ↗First posted 3 Jul 2025cs.CL cs.AIcited by 229

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

Transcending human cognitive limitations represents a critical frontier in LLM training. Proprietary agentic systems like DeepResearch have demonstrated superhuman capabilities on extremely complex information-seeking benchmarks such as BrowseComp, a feat previously unattainable. We posit that their success hinges on a sophisticated reasoning pattern absent in open-source models: the ability to systematically reduce extreme uncertainty when navigating vast information landscapes. Based on this insight, we introduce WebSailor, a complete post-training methodology designed to instill this crucial capability. Our approach involves generating novel, high-uncertainty tasks through structured sampling and information obfuscation, RFT cold start, and an efficient agentic RL training algorithm, Duplicating Sampling Policy Optimization (DUPO). With this integrated pipeline, WebSailor significantly outperforms all opensource agents in complex information-seeking tasks, matching proprietary agents' performance and closing the capability gap.

Key facts. Cited by 229 papers in the map (data through 14 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 43 of its 53 references that are in the map. First posted on arXiv 3 Jul 2025.

Builds on

43 of 53 references.

References · 43

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  6. 2024Qwen2.5 Technical Reportcited by 4,942
  7. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  8. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  9. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  10. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  11. 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,868
  12. 2016Concrete Problems in AI Safetycited by 1,642
31 more
  1. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  2. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,282
  3. 2023GAIA: a benchmark for General AI Assistantscited by 1,128
  4. 2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?cited by 982
  5. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 916
  6. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  7. 2025BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agentscited by 647
  8. 2025LIMO: Less is More for Reasoningcited by 498
  9. 2025Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Modelcited by 461
  10. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Modelscited by 433
  11. 2024Measuring short-form factuality in large language modelscited by 373
  12. 2019Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networkscited by 368
  13. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learningcited by 336
  14. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 335
  15. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 286
  16. 2023FireAct: Toward Language Agent Fine-tuningcited by 256
  17. 2025SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Modelscited by 212
  18. 2025WebDancer: Towards Autonomous Information Seeking Agencycited by 188
  19. 2025BrowseComp-ZH: Benchmarking Web Browsing Ability of Large Language Models in Chinesecited by 145
  20. 2024Leave No Document Behind: Benchmarking Long-Context LLMs with Extended Multi-Doc QAcited by 122
  21. 2024Large Language Models Must Be Taught to Know What They Don't Knowcited by 100
  22. 2022Overview frequency principle/spectral bias in deep learningcited by 69
  23. 2024LLM-based Multi-Agent Reinforcement Learning: Current and Future Directionscited by 67
  24. 2023Compositional Generalization from First Principlescited by 64
  25. 2025All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuningcited by 59
  26. 2023Calibrating LLM-Based Evaluatorcited by 53
  27. 2025EvolveSearch: An Iterative Self-Evolving Search Agentcited by 43
  28. 2025LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs -- No Silver Bullet for LC or RAG Routingcited by 29
  29. 2025Climbing the Ladder of Reasoning: What LLMs Can-and Still Can't-Solve after SFT?cited by 21
  30. 2025An overview of condensation phenomenon in deep learningcited by 17
  31. 2025Marco-o1 v2: Towards Widening The Distillation Bottleneck for Reasoning Modelscited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 86; 2026: 143.

229 papers.

Sort:

By area:

  1. 2025Memory in the Age of AI Agentscited by 303
  2. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  3. 2025BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agentcited by 200
  4. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 180
  5. 2025Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RLcited by 150
  6. 2025ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarizationcited by 141
  7. 2025Memento: Fine-tuning LLM Agents without Fine-tuning LLMscited by 127
  8. 2025WebShaper: Agentically Data Synthesizing via Information-Seeking Formalizationcited by 121
  9. 2025Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Searchcited by 120
  10. 2025WebWatcher: Breaking New Frontier of Vision-Language Deep Research Agentcited by 119
  11. 2025AgentFold: Long-Horizon Web Agents with Proactive Context Managementcited by 103
  12. 2025VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Usecited by 96
38 more
  1. 2025WebExplorer: Explore and Evolve for Training Long-Horizon Web Agentscited by 92
  2. 2025MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scalingcited by 81
  3. 2025Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RLcited by 80
  4. 2025WideSearch: Benchmarking Agentic Broad Info-Seekingcited by 69
  5. 2025WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agentscited by 68
  6. 2025Tongyi DeepResearch Technical Reportcited by 66
  7. 2026Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learningcited by 66
  8. 2025A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challengescited by 62
  9. 2025WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learningcited by 61
  10. 2025Towards General Agentic Intelligence via Environment Scalingcited by 54
  11. 2026Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligencecited by 54
  12. 2026A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agentscited by 50
  13. 2026REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agentscited by 49
  14. 2025SkyRL-Agent: Efficient RL Training for Multi-turn LLM Agentcited by 48
  15. 2025ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestrationcited by 47
  16. 2026Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agentscited by 47
  17. 2025WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Researchcited by 46
  18. 2025Budget-Aware Tool Use Enables Effective Agent Scalingcited by 44
  19. 2025MM-BrowseComp: A Comprehensive Benchmark for Multimodal Browsing Agentscited by 42
  20. 2025Scaling Agents via Continual Pre-trainingcited by 40
  21. 2026OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Datacited by 39
  22. 2025Understanding Tool-Integrated Reasoningcited by 38
  23. 2026Vision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Modelscited by 37
  24. 2026OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesiscited by 37
  25. 2025Search Self-play: Pushing the Frontier of Agent Capability without Supervisioncited by 34
  26. 2026Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Modelscited by 34
  27. 2025Demystifying Reinforcement Learning in Agentic Reasoningcited by 33
  28. 2025SFR-DeepResearch: Towards Effective Reinforcement Learning for Autonomously Reasoning Single Agentscited by 31
  29. 2025MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learningcited by 27
  30. 2026WebXSkill: Skill Learning for Autonomous Web Agentscited by 24
  31. 2025A Survey on Agentic Multimodal Large Language Modelscited by 23
  32. 2025IterResearch: Rethinking Long-Horizon Agents with Interaction Scalingcited by 23
  33. 2025MedResearcher-R1: Expert-Level Medical Deep Researcher via A Knowledge-Informed Trajectory Synthesis Frameworkcited by 22
  34. 2025A Comprehensive Survey on Reinforcement Learning-based Agentic Search: Foundations, Roles, Optimizations, Evaluations, and Applicationscited by 22
  35. 2026Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verificationcited by 22
  36. 2025DeepWideSearch: Benchmarking Depth and Width in Agentic Information Seekingcited by 21
  37. 2026OpenSearch-VL: An Open Recipe for Frontier Multimodal Search Agentscited by 21
  38. 2025Enterprise Deep Research: Steerable Multi-Agent Deep Research for Enterprise Analyticscited by 20

50 most cited of 229 All 229 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025WebCoT: Enhancing Web Agent Reasoning by Reconstructing Chain-of-Thought in Reflection, Branching, and Rollback
  2. 2026WebFactory: Automated Compression of Foundational Language Intelligence into Grounded Web Agents
  3. 2024Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents
  4. 2026WebUncertainty: Dual-Level Uncertainty Driven Planning and Reasoning For Autonomous Web Agent
  5. 2025Mind2Web 2: Evaluating Agentic Search with Agent-as-a-Judge
  6. 2026Web2BigTable: A Bi-Level Multi-Agent LLM System for Internet-Scale Information Search and Extraction
  7. 2023Mind2Web: Towards a Generalist Agent for the Web
  8. 2025DSPO: Stable and Efficient Policy Optimization for Agentic Search and Reasoning