Beyond Next-Screen Prediction: The Decision Value of GUI World Models

Separates next-screen prediction from action evaluation in GUI lookahead and trains Direct Action Ranking on reference actions, comparing prediction-augmented scoring under matched offline supervision.

AI-assisted editorial review against a primary source. This is a catalog annotation, not an independent reproduction or endorsement.

What is this resource?

Resource type

Research papers

Environment

android, web

Topics

evaluation, world-models, action-prediction, imitation-learning

Record publication date

2026-10-17

Evidence retrieved

2026-10-20T13:12:51.579373Z

Resource ID

arxiv:2610.24372

Recorded paper version

arXiv v1

Version updated: 2026-10-17T02:03:20Z.

This identifies the retained arXiv metadata snapshot; newly observed changes require separate review.

Authors

Sizhe Tang, Hanhan Zhou, Jeffrey Svajlenko, Rashmi Gangadharaiah

Cite this paper

Download BibTeX · Download CSL JSON

Citations preserve the selected source edition, its authors and recorded date precision. Unrecorded venues and identifiers are omitted.

Explore related topics

Scope and limitations

The matched augmentation experiment measures offline ranking; AndroidWorld task-success comparisons use differently trained evaluators and do not isolate prediction value. A separate web evaluator is trained on Mind2Web, and web results do not uniformly favor direct scoring. The main study concerns one-step selection. No independent reproduction.

Primary evidence and verification

Source retrieval is separate from experiment reproduction. Use the original publication or publisher page when citing claims about the resource.

1. arxiv api

   Retrieved 2026-10-20T12:58:15.922053Z. Supports: name, authors, published_at, updated_at, identifiers.arxiv, details.arxiv_snapshot.

   Content fingerprint

   b379e14eefa272f62641224fe1d839b7dfbc414d957124b8abeb103444a843ef
2. editorial primary review

   Retrieved 2026-10-20T13:12:51.579373Z. Supports: name, kind, summary, environments, topics, scope, notes.

   Content fingerprint

   9dff18935a1faa7ec16366bac6a6e9e41f0381d561568c3aba39fe783efcfbd3

Suggest a correction

Open a source-backed correction. Include this record ID and the original source. Read our editorial policy and coverage boundaries.