Efficient Agent Training for Computer Use

PC Agent-E augments 312 human computer-use trajectories with alternative action decisions synthesized by Claude 3.7 Sonnet. It studies training with this combination and introduces WindowsAgentArena-V2 as an updated evaluation benchmark.

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

windows, desktop

Topics

data-generation, evaluation

Record publication date

2025-05-20

Evidence retrieved

2026-10-16T06:36:54Z

Resource ID

arxiv:2505.13909

Recorded paper version

arXiv v2

Version updated: 2026-03-03T03:55:55Z.

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

Authors

Yanheng He, Jiahe Jin, Pengfei Liu

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.

Connected assets

These links come from reviewed resource records. A documentation link does not imply a training-data or evaluation dependency.

• Documented resource: PC-Agent-E training and deployment · Frameworks · relationship evidence
• Documented resource: WindowsAgentArena-V2 · Benchmarks · relationship evidence
• Documented resource: PC-Agent-E checkpoint · Models and checkpoints · relationship evidence
• Documented resource: PC-Agent-E training data · Datasets · relationship evidence

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 2026-10-16T06:36:54Z. Supports: name, authors, published_at, updated_at, identifiers.arxiv, details.arxiv_snapshot.

   Content fingerprint

   4953a8e0f029016aef668bce98cec4d2873a7ae49c1fe33c4311fa5eb8d24089
2. editorial primary review

   Retrieved 2026-10-16T06:36:54Z. Supports: name, kind, summary, environments, topics, scope, relations.

   Content fingerprint

   4953a8e0f029016aef668bce98cec4d2873a7ae49c1fe33c4311fa5eb8d24089

Suggest a correction

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