II.
Benchmark overview
Reference · livebenchmark:mind2web-2
Mind2Web 2 overview
Successor of Mind2Web for evaluating generalist web agents across long-horizon, multi-site, real-world tasks with judge-based outcome verification.
Attributes
displayName
Mind2Web 2
homepageUrl
kind
web-agent
targetsKind
AgentVersion
description
Successor of Mind2Web for evaluating generalist web agents across
long-horizon, multi-site, real-world tasks with judge-based outcome
verification.
Outgoing edges
applies_to1
- domain:web-development·DomainWeb Development
covers2
- skill-area:web-action-grounding·SkillAreaWeb Action Grounding
- skill-area:browser-automation·SkillAreaBrowser Automation
Incoming edges
None.