WebArena: A Realistic Web Environment for Building Autonomous Agents
Introduces WebArena, a self-hosted sandbox with five realistic websites (e-commerce, social, dev, content management, maps) and 812 high-level natural-language tasks scored by execution correctness.
- Publisher
- Carnegie Mellon University
- Year
- 2024
- Venue
- ICLR
- Authors
- 13
- Hosting
- External sourcelicense unknown
Cite
Notes
Only stored in your browser.
Introduces 1 artifact - 1 eval
TL;DR
Semantic Scholar
This paper builds an environment for language-guided agents that is highly realistic and reproducible, and creates an environment with fully functional websites from four common domains: e-commerce, social forum discussions, collaborative software development, and content management.
Artifacts
1Evals