AppVLM: A Lightweight Vision Language Model for Online App Control

AppVLM first learns app actions from AndroidControl and then refines its policy through data collected in AndroidWorld. The study compares offline action prediction and online task completion for a lightweight vision-language smartphone agent.

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

Topics

mobile-navigation, imitation-learning, data-generation, efficiency

Record publication date

2025-02-10

Evidence retrieved

2026-10-16T06:31:38Z

Resource ID

arxiv:2502.06395

Recorded paper version

arXiv v1

Version updated: 2025-02-10T12:32:21Z.

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

Authors

Georgios Papoudakis, Thomas Coste, Zhihao Wu, Jianye Hao, Jun Wang, Kun Shao

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 annotation and scope were reviewed against the primary paper abstract. Experimental results and released artifacts have not been independently reproduced.

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 2026-10-16T06:31:38Z. Supports: name, authors, published_at, updated_at, identifiers.arxiv, details.arxiv_snapshot.

   Content fingerprint

   e96b86015c763c780e0430a8d0867b96cae947f50e888b3441c85abcaa42de78
2. editorial primary review

   Retrieved 2026-10-16T06:31:38Z. Supports: name, kind, summary, environments, topics, scope.

   Content fingerprint

   e96b86015c763c780e0430a8d0867b96cae947f50e888b3441c85abcaa42de78

Suggest a correction

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