WebWorld: A Large-Scale World Model for Web Agent Training

• URL: http://arxiv.org/abs/2602.14721v1
• Date: Mon, 16 Feb 2026 13:06:49 GMT
• Title: WebWorld: A Large-Scale World Model for Web Agent Training
• Abstract summary: We introduce textbfWebWorld series, the first open-web simulator trained at scale.<n>WebWorld trains on 1M+ open-web interactions, supporting reasoning, multi-format data, and long-horizon simulations of 30+ steps.<n>We show that Qwen3-14B trained on WebWorld-synthesized trajectories improves by +9.2% on WebArena, reaching performance comparable to GPT-4o.
• Score: 59.59257499344279
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Web agents require massive trajectories to generalize, yet real-world training is constrained by network latency, rate limits, and safety risks. We introduce \textbf{WebWorld} series, the first open-web simulator trained at scale. While existing simulators are restricted to closed environments with thousands of trajectories, WebWorld leverages a scalable data pipeline to train on 1M+ open-web interactions, supporting reasoning, multi-format data, and long-horizon simulations of 30+ steps. For intrinsic evaluation, we introduce WebWorld-Bench with dual metrics spanning nine dimensions, where WebWorld achieves simulation performance comparable to Gemini-3-Pro. For extrinsic evaluation, Qwen3-14B trained on WebWorld-synthesized trajectories improves by +9.2\% on WebArena, reaching performance comparable to GPT-4o. WebWorld enables effective inference-time search, outperforming GPT-5 as a world model. Beyond web simulation, WebWorld exhibits cross-domain generalization to code, GUI, and game environments, providing a replicable recipe for world model construction.

Related papers

• Qwen-AgentWorld: Language World Models for General Agents [91.26917996662104]
  A world model predicts environment dynamics based on current observations and actions.<n>We develop Qwen-AgentWorld through a three-stage training pipeline.<n>Qwen-AgentWorld supports scalable and controllable simulation of thousands of real-world environments for agentic RL.
  arXiv  Detail & Related papers  (2026-06-23T13:53:55Z)
• Weblica: Scalable and Reproducible Training Environments for Visual Web Agents [5.132610127852326]
  Weblica is a framework for constructing reproducible and scalable web environments.<n>We scale RL training to thousands of diverse environments and tasks.<n>Our best model, Weblica-8B, outperforms open-weight baselines of similar size across multiple web navigation benchmarks.
  arXiv  Detail & Related papers  (2026-05-07T17:17:10Z)
• ComSim: Building Scalable Real-World Robot Data Generation via Compositional Simulation [90.4702774169675]
  We propose a novel hybrid approach called Compositional Simulation, which combines classical simulation and neural simulation.<n>Our approach utilizes a closed-loop real-sim-real data augmentation pipeline, leveraging a small amount of real-world data.<n>We train a neural simulator to transform classical simulation videos into real-world representations, improving the accuracy of policy models trained in real-world environments.
  arXiv  Detail & Related papers  (2026-04-13T12:25:45Z)
• DexWorldModel: Causal Latent World Modeling towards Automated Learning of Embodied Tasks [54.32016216994156]
  We introduce the Causal Latent World Model (CLWM), which employs DINOv3 features as generative targets to disentangle interaction semantics from visual noise.<n>To overcome memory scaling, CLWM features a Dual-State Test-Time Training (TTT) Memory that guarantees a strict $mathcalO(1)$ footprint for long-horizon tasks.<n>We present EmbodiChain, an online framework that establishes the Efficiency Law by injecting an infinite flow of physics-grounded trajectories during training.
  arXiv  Detail & Related papers  (2026-04-13T03:19:36Z)
• MolmoWeb: Open Visual Web Agent and Open Data for the Open Web [60.29597961827816]
  MolmoWebMix is a mixture of browser task demonstrations and web-GUI perception data.<n>MolmoWeb-8B is a family of fully open multimodal web agents.<n>We will release model checkpoints, training data, code, and a unified evaluation harness to enable open research on web agents.
  arXiv  Detail & Related papers  (2026-04-09T17:54:02Z)
• WebChain: A Large-Scale Human-Annotated Dataset of Real-World Web Interaction Traces [5.150606279179606]
  WebChain is the largest open-source dataset of human-annotated trajectories on real-world websites.<n>Our work provides the data and insights necessary to build and rigorously evaluate the next generation of scalable web agents.
  arXiv  Detail & Related papers  (2026-03-05T15:37:34Z)
• AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State Machines [43.251303612671194]
  AutoWebWorld is a framework for synthesizing controllable and verifiable web environments.<n>Unlike real websites, where state transitions are implicit, AutoWebWorld explicitly defines all states, actions, and transition rules.<n>It generates over 11,663 verified trajectories from 29 diverse web environments at only $0.04 per trajectory.
  arXiv  Detail & Related papers  (2026-02-15T20:03:19Z)
• Web World Models [60.208836336654315]
  We introduce the Web World Model (WWM), a middle ground where world state and physics'' are implemented in ordinary web code.<n>We build a suite of WWMs on a realistic web stack, including an infinite travel atlas grounded in real geography, fictional galaxy explorers, web-scale encyclopedic and narrative worlds, and simulation- and game-like environments.<n>Our results suggest that web stacks themselves can serve as a scalable substrate for world models, enabling controllable yet open-ended environments.
  arXiv  Detail & Related papers  (2025-12-29T18:31:45Z)
• GigaWorld-0: World Models as Data Engine to Empower Embodied AI [42.7374586431707]
  We present GigaWorld-0, a unified world model framework designed explicitly as a data engine for Vision-Language-Action learning.<n>GigaWorld-0 integrates two synergistic components: GigaWorld-0-Video, which leverages large-scale video generation to produce diverse, texture-rich, and temporally coherent embodied sequences.<n>GigaWorld-0-3D combines 3D generative modeling, 3D Gaussian Splatting reconstruction, physically differentiable system identification, and executable motion planning to ensure geometric consistency and physical realism.
  arXiv  Detail & Related papers  (2025-11-25T03:00:42Z)
• LLMs as Scalable, General-Purpose Simulators For Evolving Digital Agent Training [55.72784274656801]
  We introduce a scalable paradigm that generates structured UI states and transitions to synthesize training trajectories at scale.<n>Our paradigm integrates a digital world simulator for diverse UI states, a guided rollout process for coherent exploration, and a trajectory wrapper.<n>Experiments on WebArena and AndroidWorld show that UI-Simulator rivals or surpasses open-source agents trained on real UIs with significantly better robustness.
  arXiv  Detail & Related papers  (2025-10-16T17:59:38Z)
• Trajectory World Models for Heterogeneous Environments [67.27233466954814]
  Heterogeneity in sensors and actuators across environments poses a significant challenge to building large-scale pre-trained world models.<n>We introduce UniTraj, a unified dataset comprising over one million trajectories from 80 environments, designed to scale data while preserving critical diversity.<n>We also propose TrajWorld, a novel architecture capable of handling varying sensor and actuator information and capturing environment dynamics in-context.
  arXiv  Detail & Related papers  (2025-02-03T13:59:08Z)
• Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents [22.608219492706876]
  We propose a model-based planning framework for web agents that employs a world model to simulate and deliberate over the outcome of each candidate action before committing to one.<n> Empirical results demonstrate that WebDreamer achieves substantial performance improvements over reactive baselines.<n>Our trained world model, Dreamer-7B, performs comparable to GPT-4o, highlighting the potential of specialized world models for efficient and effective planning in complex web environments.
  arXiv  Detail & Related papers  (2024-11-10T18:50:51Z)
• WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models [65.18602126334716]
  Existing web agents typically only handle one input modality and are evaluated only in simplified web simulators or static web snapshots. We introduce WebVoyager, an innovative Large Multimodal Model (LMM) powered web agent that can complete user instructions end-to-end by interacting with real-world websites. We show that WebVoyager achieves a 59.1% task success rate on our benchmark, significantly surpassing the performance of both GPT-4 (All Tools) and the WebVoyager (text-only) setups.
  arXiv  Detail & Related papers  (2024-01-25T03:33:18Z)
• Learning Interactive Real-World Simulators [96.5991333400566]
  We explore the possibility of learning a universal simulator of real-world interaction through generative modeling. We use the simulator to train both high-level vision-language policies and low-level reinforcement learning policies. Video captioning models can benefit from training with simulated experience, opening up even wider applications.
  arXiv  Detail & Related papers  (2023-10-09T19:42:22Z)

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.