Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks

Qian WAN; Jiannan LI; Huanchen WANG; Zhicong LU

doi:10.1145/3757497

Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks

Qian WAN, Jiannan LI, Huanchen WANG, Zhicong LU

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

Prewriting is the process of generating and organising ideas before a first draft. It consists of a combination of informal, iterative, and semi-structured strategies such as visual diagramming, which poses a challenge for collaborating with large language models (LLMs) in a turn-taking conversational manner. We present Polymind, a visual diagramming tool that leverages multiple LLM-powered agents to support prewriting. The system features a parallel collaboration workflow in place of the turn-taking conversational interactions. It defines multiple "microtasks" to simulate group collaboration scenarios such as collaborative writing and group brainstorming. Instead of repetitively prompting a chatbot for various purposes, Polymind enables users to orchestrate multiple microtasks simultaneously. Users can configure and delegate customised microtasks, and manage their microtasks by specifying task requirements and toggling visibility and initiative. Our evaluation revealed that, compared to ChatGPT, users had more customizability over collaboration with Polymind, and were thus able to quickly expand personalised writing ideas during prewriting.

© 2025 Copyright held by the owner/author(s).

Original language	English
Title of host publication	Proceedings of the ACM on Human-Computer Interaction
Editors	Jeff Nichols
Publisher	Association for Computing Machinery
Volume	9
DOIs	https://doi.org/10.1145/3757497
Publication status	Published - Oct 2025
Event	28th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2025) - Grieghallen Conference Center, Bergen, Norway Duration: 18 Oct 2025 → 22 Oct 2025 https://cscw.acm.org/2025/

Publication series

Name	Proceedings of the ACM on Human-Computer Interaction
Publisher	Association for Computing Machinery
ISSN (Print)	2573-0142

Conference

Conference	28th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2025)
Abbreviated title	CSCW2025
Place	Norway
City	Bergen
Period	18/10/25 → 22/10/25
Internet address	https://cscw.acm.org/2025/

Bibliographical note

Research Unit(s) information for this publication is provided by the author(s) concerned.

Research Keywords

creativity support
diagramming
human-ai collaboration
microtasking
prewriting

Access Output

10.1145/3757497

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

WAN, Q., LI, J., WANG, H., & LU, Z. (2025). Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks. In J. Nichols (Ed.), Proceedings of the ACM on Human-Computer Interaction (Vol. 9). Article CSCW316 (Proceedings of the ACM on Human-Computer Interaction). Association for Computing Machinery. https://doi.org/10.1145/3757497

@inproceedings{da8256b0f6e4409f91c7a24deb20126e,

title = "Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks",

abstract = " Prewriting is the process of generating and organising ideas before a first draft. It consists of a combination of informal, iterative, and semi-structured strategies such as visual diagramming, which poses a challenge for collaborating with large language models (LLMs) in a turn-taking conversational manner. We present Polymind, a visual diagramming tool that leverages multiple LLM-powered agents to support prewriting. The system features a parallel collaboration workflow in place of the turn-taking conversational interactions. It defines multiple {"}microtasks{"} to simulate group collaboration scenarios such as collaborative writing and group brainstorming. Instead of repetitively prompting a chatbot for various purposes, Polymind enables users to orchestrate multiple microtasks simultaneously. Users can configure and delegate customised microtasks, and manage their microtasks by specifying task requirements and toggling visibility and initiative. Our evaluation revealed that, compared to ChatGPT, users had more customizability over collaboration with Polymind, and were thus able to quickly expand personalised writing ideas during prewriting. {\textcopyright} 2025 Copyright held by the owner/author(s).",

keywords = "creativity support, diagramming, human-ai collaboration, microtasking, prewriting",

author = "Qian WAN and Jiannan LI and Huanchen WANG and Zhicong LU",

note = "Research Unit(s) information for this publication is provided by the author(s) concerned. ; 28th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2025), CSCW2025 ; Conference date: 18-10-2025 Through 22-10-2025",

year = "2025",

month = oct,

doi = "10.1145/3757497",

language = "English",

volume = "9",

series = "Proceedings of the ACM on Human-Computer Interaction",

publisher = "Association for Computing Machinery",

editor = "Jeff Nichols",

booktitle = "Proceedings of the ACM on Human-Computer Interaction",

address = "United States",

url = "https://cscw.acm.org/2025/",

}

WAN, Q, LI, J, WANG, H & LU, Z 2025, Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks. in J Nichols (ed.), Proceedings of the ACM on Human-Computer Interaction. vol. 9, CSCW316, Proceedings of the ACM on Human-Computer Interaction, Association for Computing Machinery, 28th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2025), Bergen, Norway, 18/10/25. https://doi.org/10.1145/3757497

Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks. / WAN, Qian; LI, Jiannan; WANG, Huanchen et al.
Proceedings of the ACM on Human-Computer Interaction. ed. / Jeff Nichols. Vol. 9 Association for Computing Machinery, 2025. CSCW316 (Proceedings of the ACM on Human-Computer Interaction).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Polymind

T2 - 28th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2025)

AU - WAN, Qian

AU - LI, Jiannan

AU - WANG, Huanchen

AU - LU, Zhicong

N1 - Research Unit(s) information for this publication is provided by the author(s) concerned.

PY - 2025/10

Y1 - 2025/10

N2 - Prewriting is the process of generating and organising ideas before a first draft. It consists of a combination of informal, iterative, and semi-structured strategies such as visual diagramming, which poses a challenge for collaborating with large language models (LLMs) in a turn-taking conversational manner. We present Polymind, a visual diagramming tool that leverages multiple LLM-powered agents to support prewriting. The system features a parallel collaboration workflow in place of the turn-taking conversational interactions. It defines multiple "microtasks" to simulate group collaboration scenarios such as collaborative writing and group brainstorming. Instead of repetitively prompting a chatbot for various purposes, Polymind enables users to orchestrate multiple microtasks simultaneously. Users can configure and delegate customised microtasks, and manage their microtasks by specifying task requirements and toggling visibility and initiative. Our evaluation revealed that, compared to ChatGPT, users had more customizability over collaboration with Polymind, and were thus able to quickly expand personalised writing ideas during prewriting. © 2025 Copyright held by the owner/author(s).

AB - Prewriting is the process of generating and organising ideas before a first draft. It consists of a combination of informal, iterative, and semi-structured strategies such as visual diagramming, which poses a challenge for collaborating with large language models (LLMs) in a turn-taking conversational manner. We present Polymind, a visual diagramming tool that leverages multiple LLM-powered agents to support prewriting. The system features a parallel collaboration workflow in place of the turn-taking conversational interactions. It defines multiple "microtasks" to simulate group collaboration scenarios such as collaborative writing and group brainstorming. Instead of repetitively prompting a chatbot for various purposes, Polymind enables users to orchestrate multiple microtasks simultaneously. Users can configure and delegate customised microtasks, and manage their microtasks by specifying task requirements and toggling visibility and initiative. Our evaluation revealed that, compared to ChatGPT, users had more customizability over collaboration with Polymind, and were thus able to quickly expand personalised writing ideas during prewriting. © 2025 Copyright held by the owner/author(s).

KW - creativity support

KW - diagramming

KW - human-ai collaboration

KW - microtasking

KW - prewriting

UR - http://www.scopus.com/inward/record.url?scp=105019240003&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-105019240003&origin=recordpage

U2 - 10.1145/3757497

DO - 10.1145/3757497

M3 - RGC 32 - Refereed conference paper (with host publication)

VL - 9

T3 - Proceedings of the ACM on Human-Computer Interaction

BT - Proceedings of the ACM on Human-Computer Interaction

A2 - Nichols, Jeff

PB - Association for Computing Machinery

Y2 - 18 October 2025 through 22 October 2025

ER -

Polymind: Parallel Visual Diagramming with Large Language Models to Support Prewriting Through Microtasks

Abstract

Publication series

Conference

Bibliographical note

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this