WebHarbor: Docking Real Websites for Evolving GUI Agent Environments
GUI web agents require realistic and interactive environments that remain stable during both evaluation and training. Live website evaluation suffers from changing content, access restrictions, and limited coverage of authenticated workflows, while existing synthetic environments for training often lack functional depth and visual fidelity. We propose WebHarbor, a suite of 153 executable and…
We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.
