Mind2Web-SC

A GuardAgent safety eval: generate and execute code that grants or denies a SeeAct web action under six user-constraint rules.

Also known as: Mind2Web SC, Mind2Web Safety Control

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorysafety
Subcategoryweb-agent guardrails (SeeAct + user constraints)
Page statusactive
Metricaccuracy (comprehensive correctness: decision plus violation on DENY)
Directionhigher_is_better
Unit%
Dataset size200
Dataset licenceinspect_evals MIT; GuardAgent paper HTML states CC BY 4.0; original data also hosted at guardagent/dataset
PublisherUniversity of Georgia, University of Chicago, UIUC, UT Austin, UC Berkeley, Emory, Virtue AI (GuardAgent authors)

What it measures

Mind2Web-SC tests a guard model, not a web navigator. The model sees a user task from Mind2Web, a synthetic user profile (age plus boolean flags for membership, vaccine, driver's licence, and domestic status), and a SeeAct agent output. It must decompose the applicable safety rule, write Python guardrail code, and decide GRANT or DENY. On DENY it must also name the violated rule. English text plus executed code. It is not the [Mind2Web](mind2web.md) action-prediction task.

Task format

Few-shot code generation (inspect default num_shots=3) inside a Docker sandbox. inspect_evals asks for task decomposition and guardrail code, runs the code, then parses "Action granted" / "Action denied" and an optional "Violation:" line. Paper metrics include label precision/recall and comprehensive control accuracy; inspect reports accuracy() plus accuracy grouped by domain.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub