Multi-Agent Based Message Generation and Delivery for Personalized Environmental Awareness Promotion in Urban City

ACL ARR 2025 July Submission110 Authors

Abstract: With growing concerns about climate change, effectively promoting pro-environmental behavior becomes a pressing societal challenge. While traditional publicity strategies are usually general, targeting diverse citizen profiles and behavioral motivations with different strategies, sounds to be much more appropriate. In this paper, we propose MA-MGD, a Multi-Agent Based Message Generation and Delivery Framework, which use multi-agent system to generate personalized publicity strategy for citizen with different profiles and behavioral motivations, in a context of environment protection promotion. The system consists of citizen profiling, simulation testing and iterative feedback to promote environment-friendly living styles and low-carbon behavioral changes. Deployed on the AgentSociety platform, the system targets 200 virtual citizens in a simulated Beijing environment and dynamically delivers personalized messages and posters through multi-agent collaboration, to promote environment protection. Experimental results demonstrate that, compared to static or template-based methods, MA-MGD significantly improves message rationality, carbon awareness, and low-carbon travel intent. Our findings highlight the potential of LLM-based multi-agent frameworks in enabling cost-effective, adaptive, and behaviorally impactful environmental interventions.

Paper Type: Long

Research Area: Computational Social Science and Cultural Analytics

Research Area Keywords: human behavior analysis, NLP tools for social analysis, Psycho-demographic trait prediction

Contribution Types: NLP engineering experiment, Data analysis, Surveys

Languages Studied: Chinese, English

Reassignment Request Area Chair: This is not a resubmission

Reassignment Request Reviewers: This is not a resubmission

A1 Limitations Section: This paper has a limitations section.

A2 Potential Risks: N/A

B Use Or Create Scientific Artifacts: No

B1 Cite Creators Of Artifacts: N/A

B1 Elaboration: N/A

B2 Discuss The License For Artifacts: N/A

B2 Elaboration: N/A

B3 Artifact Use Consistent With Intended Use: N/A

B3 Elaboration: N/A

B4 Data Contains Personally Identifying Info Or Offensive Content: N/A

B5 Documentation Of Artifacts: N/A

B6 Statistics For Data: N/A

B6 Elaboration: N/A

C Computational Experiments: No

C1 Model Size And Budget: N/A

C1 Elaboration: N/A

C2 Experimental Setup And Hyperparameters: N/A

C2 Elaboration: N/A

C3 Descriptive Statistics: N/A

C3 Elaboration: N/A

C4 Parameters For Packages: N/A

D Human Subjects Including Annotators: No

D1 Instructions Given To Participants: N/A

D2 Recruitment And Payment: N/A

D3 Data Consent: N/A

D4 Ethics Review Board Approval: N/A

D5 Characteristics Of Annotators: N/A

E Ai Assistants In Research Or Writing: Yes

E1 Information About Use Of Ai Assistants: N/A

Author Submission Checklist: yes

Submission Number: 110