Explorer: Scaling Exploration-driven Web Trajectory Synthesis for Multimodal Web Agents
Vardaan Pahuja , Yadong Lu , Corby Rosset , Boyu Gou , Arindam Mitra , Spencer Whitehead , Yu Su , Ahmed Hassan Awadallah
- 🏛 Institutions
- MSR , Redmond , OSU
- 📅 Date
- July 2025
- 📑 Publisher
- Findings of ACL 2025
- 💻 Env
- Web
- 🔑 Keywords
TLDR
This paper targets the shortage of large, diverse web-agent trajectories by synthesizing a 94K-trajectory multimodal dataset through scalable exploration and refinement over 49K URLs. Training on this dataset yields a strong web agent, Explorer, and shows that data scaling is a major driver of web-agent performance.
Related papers (24)
- Video2GUI: Synthesizing Large-Scale Interaction Trajectories for Generalized GUI Agent PretrainingMay 14, 2026 · arXiv
- OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task SynthesisDecember 27, 2024 · ACL 2025
- On the Effects of Data Scale on UI Control AgentsJune 6, 2024 · NeurIPS 2024 Datasets and Benchmarks Track
- WebForge: Breaking the Realism-Reproducibility-Scalability Trilemma in Browser Agent BenchmarkApril 13, 2026 · arXiv
- MolmoWeb: Open Visual Web Agent and Open Data for the Open WebApril 9, 2026 · arXiv
- WebArena-Infinity: Generating Browser Environments with Verifiable Tasks at ScaleMarch 2026 · Blog Post
- HATS: Hardness-Aware Trajectory Synthesis for GUI AgentsMarch 12, 2026 · CVPR 2026
- WebChain: A Large-Scale Human-Annotated Dataset of Real-World Web Interaction TracesMarch 5, 2026 · arXiv
- Modeling Distinct Human Interaction in Web AgentsFebruary 19, 2026 · arXiv
- WebWorld: A Large-Scale World Model for Web Agent TrainingFebruary 16, 2026 · arXiv
- OpenCUA: Open Foundations for Computer-Use AgentsAugust 12, 2025 · NeurIPS 2025 (Spotlight)
- Web-CogReasoner: Towards Knowledge-Induced Cognitive Reasoning for Web AgentsAugust 3, 2025 · ICLR 2026 (Poster)
- Go-Browse: Training Web Agents with Structured ExplorationJune 4, 2025 · ICLR 2026 (Poster)
- Web-Shepherd: Advancing PRMs for Reinforcing Web AgentsMay 21, 2025 · NeurIPS 2025 (Spotlight)
- ScaleTrack: Scaling and back-tracking Automated GUI AgentsMay 1, 2025 · arXiv
- RealWebAssist: A Benchmark for Long-Horizon Web Assistance with Real-World UsersApril 14, 2025 · AAAI 2026
- AgentRewardBench: Evaluating Automatic Evaluations of Web Agent TrajectoriesApril 11, 2025 · COLM 2025
- SpiritSight Agent: Advanced GUI Agent with One LookMarch 5, 2025 · CVPR 2025 (Poster)
- ShowUI: One Vision-Language-Action Model for GUI Visual AgentNovember 26, 2024 · CVPR 2025 (Poster)
- OS-ATLAS: A Foundation Action Model for Generalist GUI AgentsOctober 30, 2024 · ICLR 2025 (Spotlight)
- Harnessing Webpage UIs for Text-Rich Visual UnderstandingOctober 17, 2024 · ICLR 2025 (Poster)
- Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI AgentsOctober 7, 2024 · ICLR 2025 (Oral)
- Synatra: Turning Indirect Knowledge into Direct Demonstrations for Digital Agents at ScaleSeptember 24, 2024 · NeurIPS 2024
- WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work TasksJuly 7, 2024 · NeurIPS 2024 Datasets and Benchmarks Track (Poster)