SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.07945 · IJCAI · 2024

ScreenAgent: A Vision Language Model-driven Computer Control Agent

Shiqi Wang, Qi Wang, Yi Chang, Jindong Li, Runliang Niu, Yali Fu, Xiyu Hu, Xueyuan Leng, He Kong

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 4 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2402.07945.json · badge: /api/ran/2402.07945.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
niuzaisheng/ScreenAgent canonical (classifier label)
classifier basis: census
6 ran · 3 no recorded run
FunctionStatusWhere it lives
compute_bleu Ran
ran; our contract draft was wrong, not the code
niuzaisheng/ScreenAgent/client/bleu.py
pointer only (licence: NONE); the stored copy get_code serves, from WangXFng/RDRec, is pointer only (licence: NONE) · get_code("49289a7f94c227bd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, niuzaisheng/ScreenAgent when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

data_collator Ran
ran; our contract draft was wrong, not the code
niuzaisheng/ScreenAgent/train/finetune_ScreenAgent.py
pointer only (licence: NOASSERTION) · get_code("ad5ba95300a1cc7a")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_image_from_base64 Ran
ran, honoured the contract we drafted
niuzaisheng/ScreenAgent/train/cogagent_model_worker.py
pointer only (licence: NOASSERTION) · get_code("6493522dff8f8fa5")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pack_ard Ran
ran; our contract draft was wrong, not the code
niuzaisheng/ScreenAgent/client/asyncvnc.py
pointer only (licence: NOASSERTION) · get_code("625f66bc3b7b4b5b")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

read_int Ran
ran on a synthesized input
niuzaisheng/ScreenAgent/client/asyncvnc.py
pointer only (licence: NOASSERTION) · get_code("6741565cee168112")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

read_text Ran
ran; our contract draft was wrong, not the code
niuzaisheng/ScreenAgent/client/asyncvnc.py
pointer only (licence: NOASSERTION) · get_code("52dc9a4327ee52bd")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

convert_mouse_button Not yet run
unverified (harvested, no recorded run)
niuzaisheng/ScreenAgent/client/base.py
pointer only (licence: NOASSERTION) · get_code("c87254327d5ba881")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

find_non_json_span_from_text Not yet run
unverified (harvested, no recorded run)
niuzaisheng/ScreenAgent/client/action.py
pointer only (licence: NOASSERTION) · get_code("3679d1592673c770")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

split_json_span_and_non_json_span_from_text Not yet run
unverified (harvested, no recorded run)
niuzaisheng/ScreenAgent/client/action.py
pointer only (licence: NOASSERTION) · get_code("6bf13d2e007c2458")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, niuzaisheng/ScreenAgent (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
compute_bleu 10 other papers
7 dated earlier, 3 later
1909.09436 (2019-09)
CodeSearchNet Challenge: Evaluating the State of Semantic Code Search
aaai_6453 (2020 (year only))
Improving Knowledge-Aware Dialogue Generation via Knowledge Base Question Answering
2206.05239 (2022-06)
StructCoder: Structure-Aware Transformer for Code Generation
and 7 more not named here
data_collator 1 other paper
1 dated earlier
2312.08914 (2023-12)
CogAgent: A Visual Language Model for GUI Agents

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Large Language Models (LLM) can invoke a variety of tools and APIs to complete complex tasks. The computer, as the most powerful and universal tool, could potentially be controlled by a trained LLM agent. Powered by the computer, we can hopefully build a more generalized agent to assist humans in various daily digital works. In this paper, we construct an environment for a Vision Language Model (VLM) agent to interact with a real computer screen. Within this environment, the agent can observe screenshots and manipulate the Graphical User Interface (GUI) by outputting mouse and keyboard actions. We also design an automated control pipeline that includes planning, acting, and reflecting phases, guiding the agent to continuously interact with the environment and complete multistep tasks. Additionally, we construct the ScreenAgent Dataset, which collects screenshots and action sequences when completing daily computer tasks. Finally, we train a model, ScreenAgent, which achieves comparable computer control capabilities to GPT-4V and demonstrated more precise UI positioning capabilities. Our attempts could inspire further research on building a generalist LLM agent. The code and more detailed information are at https://github.com/niuzaisheng/ScreenAgent. Planning:Now, your current task is "Start the presentation from the third slide", give the disassembly steps of is "Start the presentation from the third slide", please describe whether this image meets the goal in json format? And whether our mission can continue. Acting : Based on the provided screen image. You can use the mouse and keyboard. Please make output execution actions... Task Prompt: Start the presentation from the third slide click(114,360) Before After VLM Agent the task based on the state of the existing screen image.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2402.07945")
get_code_for_paper("2402.07945")
have("2402.07945")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2402.07945.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2402.07945.svg)](https://syntology.ai/paper/2402.07945)

Connect an agent — have() is free.