mlsanityData to 14 Oct 2026

NNetNav: Unsupervised Learning of Browser Agents Through Environment Interaction in the Wild

arXiv 2410.02907 ↗First posted 3 Oct 2024cs.CLcited by 59

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

We introduce NNetNav, a method for unsupervised interaction with websites that generates synthetic demonstrations for training browser agents. Given any website, NNetNav produces these demonstrations by retroactively labeling action sequences from an exploration policy. Most work on training browser agents has relied on expensive human supervision, and the limited prior work on such interaction-based techniques has failed to provide effective search through the exponentially large space of exploration. In contrast, NNetNav exploits the hierarchical structure of language instructions to make this search more tractable: Complex instructions are typically decomposable into simpler sub-tasks, allowing NNetNav to automatically prune interaction episodes when an intermediate trajectory cannot be annotated with a meaningful sub-task. textttLLama-3.1-8b finetuned on 10k NNetNav self-generated demonstrations obtains over 16% success rate on WebArena, and 35% on WebVoyager, an improvement of 15pts and 31pts respectively over zero-shot textttLLama-3.1-8b, outperforming zero-shot GPT-4 and reaching the state-of-the-art among unsupervised methods, for both benchmarks.

Key facts. Cited by 59 papers in the map (data through 14 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 26 of its 39 references that are in the map. First posted on arXiv 3 Oct 2024.

Builds on

26 of 39 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024Agent Workflow Memory11 Sep
  2. 2024Large Language Models Can Self-Improve At Web Agent Tasks30 May
  3. 2024WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?12 Mar
  4. 2024WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models25 Jan
  5. 2023WebArena: A Realistic Web Environment for Building Autonomous Agents25 Jul

Other references · 21

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  5. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  6. 2019ZeRO: Memory Optimizations Toward Training Trillion Parameter Modelscited by 1,301
  7. 2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,195
  8. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 752
  9. 2023Language Models can Solve Computer Taskscited by 372
  10. 2023Role-Play with Large Language Modelscited by 294
  11. 2018Reinforcement Learning on Web Interfaces Using Workflow-Guided Explorationcited by 284
  12. 2023AdaPlanner: Adaptive Planning from Feedback with Language Modelscited by 196
9 more
  1. 2023Guiding Pretraining in Reinforcement Learning with Large Language Modelscited by 187
  2. 2024WebLINX: Real-World Website Navigation with Multi-Turn Dialoguecited by 180
  3. 2023Multimodal Web Navigation with Instruction-Finetuned Foundation Modelscited by 160
  4. 2022A data-driven approach for learning to control computerscited by 88
  5. 2017Mapping Instructions and Visual Observations to Actions with Reinforcement Learningcited by 81
  6. 2022Improving Intrinsic Exploration with Language Abstractionscited by 49
  7. 2024Synatra: Turning Indirect Knowledge into Direct Demonstrations for Digital Agents at Scalecited by 43
  8. 2024BAGEL: Bootstrapping Agents by Guiding Exploration with Languagecited by 34
  9. 2017Guiding Reinforcement Learning Exploration Using Natural Languagecited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 23; 2026: 34.

59 papers.

Sort:

By area:

  1. 2025SWE-smith: Scaling Data for Software Engineering Agentscited by 285
  2. 2024Large Language Model-Brained GUI Agents: A Surveycited by 162
  3. 2024AgentTrek: Agent Trajectory Synthesis via Guiding Replay with Web Tutorialscited by 130
  4. 2025AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectoriescited by 108
  5. 2025WebEvolver: Enhancing Web Agent Self-Improvement with Coevolving World Modelcited by 61
  6. 2025Self-Challenging Language Model Agentscited by 59
  7. 2025OdysseyBench: Evaluating LLM Agents on Long-Horizon Complex Office Application Workflowscited by 57
  8. 2025AgentSynth: Scalable Task Generation for Generalist Computer-Use Agentscited by 46
  9. 2025Thinking vs. Doing: Agents that Reason by Scaling Test-Time Interactioncited by 34
  10. 2025Go-Browse: Training Web Agents with Structured Explorationcited by 33
  11. 2025InSTA: Towards Internet-Scale Training For Agentscited by 32
  12. 2025PolySkill: Learning Generalizable Skills Through Polymorphic Abstractioncited by 27
38 more
  1. 2026WebXSkill: Skill Learning for Autonomous Web Agentscited by 24
  2. 2026AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State Machinescited by 21
  3. 2025Test-Time Adaptation for LLM Agents via Environment Interactioncited by 20
  4. 2025How to Train Your LLM Web Agent: A Statistical Diagnosiscited by 19
  5. 2025Build the web for agents, not agents for the webcited by 18
  6. 2025LLMs as Scalable, General-Purpose Simulators For Evolving Digital Agent Trainingcited by 18
  7. 2025Agent Data Protocol: Unifying Datasets for Diverse, Effective Fine-tuning of LLM Agentscited by 17
  8. 2025DECEPTICON: How Dark Patterns Manipulate Web Agentscited by 15
  9. 2025SynthAgent: Adapting Web Agents with Synthetic Supervisioncited by 14
  10. 2026DynaWeb: Model-Based Reinforcement Learning of Web Agentscited by 14
  11. 2025Hidden Ghost Hand: Unveiling Backdoor Vulnerabilities in MLLM-Powered Mobile GUI Agentscited by 13
  12. 2026The Art of Building Verifiers for Computer Use Agentscited by 13
  13. 2026Safe and Scalable Web Agent Learning via Recreated Websitescited by 11
  14. 2026OpenMobile: Building Open Mobile Agents with Task and Trajectory Synthesiscited by 11
  15. 2025Watch and Learn: Learning to Use Computers from Online Videoscited by 9
  16. 2026OpAgent: Operator Agent for Web Navigationcited by 8
  17. 2026PABU: Progress-Aware Belief Update for Efficient LLM Agentscited by 7
  18. 2026MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agentscited by 7
  19. 2026PPT-Eval: A Benchmark for Computer-Use Agents on PowerPoint Taskscited by 7
  20. 2025Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chaincited by 6
  21. 2026WebFactory: Automated Compression of Foundational Language Intelligence into Grounded Web Agentscited by 6
  22. 2026From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AIcited by 6
  23. 2026Fara-1.5: Scalable Learning Environments for Computer Use Agentscited by 6
  24. 2025Reflection-Based Memory For Web navigation Agentscited by 5
  25. 2026On the Fragility of Self-Improving Agents: Variance, Task Order, and Underspecificationcited by 4
  26. 2026Video-Based Reward Modeling for Computer-Use Agentscited by 3
  27. 2026Supervised Fine-Tuning versus Reinforcement Learning: A Study of Post-Training Methods for Large Language Modelscited by 3
  28. 2026Structured Distillation of Web Agent Capabilities Enables Generalizationcited by 3
  29. 2025SynQuE: Estimating Synthetic Dataset Quality Without Annotationscited by 2
  30. 2025Instruction Agent: Enhancing Agent with Expert Demonstrationcited by 1
  31. 2026STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Taskscited by 1
  32. 2026EnterpriseLab: A Full-Stack Platform for developing and deploying agents in Enterprisescited by 1
  33. 2026Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directionscited by 1
  34. 2026GTA: Generating Long-Horizon Tasks for Web Agents at Scalecited by 1
  35. 2026ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Taskscited by 1
  36. 2026Software Engineering for and with GUI Agentcited by 1
  37. 2026Benchmarking LLM Judges for Mobile Agent Evaluationcited by 1
  38. 2026GUI agent: Guided Exploration of User-Sensitive Screenscited by 0

50 most cited of 59 All 59 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
  2. 2026MolmoWeb: Open Visual Web Agent and Open Data for the Open Web
  3. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents
  4. 2024OpenWebVoyager: Building Multimodal Web Agents via Iterative Real-World Exploration, Feedback and Optimization
  5. 2026SkillWeaver: Agentic Exploration over Neural Interaction Skills for Scalable Robot Data Generation
  6. 2026OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents
  7. 2025WebATLAS: An LLM Agent with Experience-Driven Memory and Action Simulation
  8. 2026SCAFFOLD: Self-Improving Web Agents via Recursive Parametric Skill Abstraction