A Real-World WebAgent with Planning, Long Context Understanding, and Program Synthesis
Izzeddin Gur, Hiroki Furuta, Austin Huang, Mustafa Safdari, Yutaka Matsuo, Douglas Eck, Aleksandra Faust
- 🏛 Institutions
- Google DeepMind, University of Tokyo
- 📅 Date
- July 24, 2023
- 📑 Publisher
- ICLR 2024 (Oral)
- 💻 Env
- Web
- 🔑 Keywords
TLDR
WebAgent is a modular real-world web agent that decomposes instructions into sub-instructions, summarizes long HTML into task-relevant snippets, and executes generated Python programs on websites. The paper pairs that agent design with HTML-T5, a long-context model for HTML planning and summarization.
Related papers
- WebATLAS: An LLM Agent with Experience-Driven Memory and Action SimulationOctober 26, 2025 · NeurIPS 2025 Workshop on Language Agents and World Models
- LiteWebAgent: The Open-Source Suite for VLM-Based Web-Agent ApplicationsMarch 4, 2025 · NAACL 2025 System Demonstrations
- Building a Stable Planner: An Extended Finite State Machine Based Planning Module for Mobile GUI AgentMay 20, 2025 · arXiv
- OpeFlo: Automated UX Evaluation via Simulated Human Web Interaction with GUI GroundingFebruary 25, 2026 · arXiv
- ColorBrowserAgent: Complex Long-Horizon Browser Agent with Adaptive Knowledge EvolutionJanuary 12, 2026 · arXiv
- Surfer 2: The Next Generation of Cross-Platform Computer Use AgentsOctober 22, 2025 · arXiv