Point of Order: Action-Aware LLM Persona Modeling for Data-Grounded Civic Deliberation

arXiv cs.CL Papers

Summary

A reproducible pipeline converts public Zoom recordings into speaker-attributed transcripts enriched with personas, topics, and action tags, then fine-tunes LLM personas on this data. Action-aware fine-tuning significantly improves persona fidelity, consistency, and deliberative responsiveness, enabling realistic civic deliberation simulations.

arXiv:2511.17813v3 Announce Type: replace Abstract: LLM-based simulations can enable controlled studies of civic deliberation, but current systems lack speaker-attributed data and methods for evaluating long-form institutional behavior. ASR transcripts typically use anonymous labels such as $Speaker\_1$, preventing models from learning stable participant behavior across meetings. We present a reproducible pipeline that converts public Zoom recordings into speaker-attributed transcripts enriched with persona profiles, topics, and pragmatic "action tags" such as $[propose\_motion]$. Using this pipeline, we release three public datasets of government deliberation (Appellate Court hearings, School Board meetings, and Municipal Council sessions) and fine-tune LLM personas on this action-aware data. We evaluate simulations along four dimensions: persona fidelity, persona consistency, institutional fidelity, and behavioral coherence. Action-aware fine-tuning cuts perplexity by 67%, doubles classifier-based persona fidelity, increases vote attempts by up to $3.6\times$, and improves deliberative responsiveness by up to 70%. Human evaluations show that simulated excerpts are often hard to distinguish from real deliberations, indicating a practical foundation for data-grounded civic simulation studies.
Original Article
View Cached Full Text

Cached at: 07/13/26, 08:00 AM

# Point of Order: Action-Aware LLM Persona Modeling for Data-Grounded Civic Deliberation
Source: [https://arxiv.org/abs/2511.17813](https://arxiv.org/abs/2511.17813)
[View PDF](https://arxiv.org/pdf/2511.17813)

> Abstract:LLM\-based simulations can enable controlled studies of civic deliberation, but current systems lack speaker\-attributed data and methods for evaluating long\-form institutional behavior\. ASR transcripts typically use anonymous labels such as $Speaker\\\_1$, preventing models from learning stable participant behavior across meetings\. We present a reproducible pipeline that converts public Zoom recordings into speaker\-attributed transcripts enriched with persona profiles, topics, and pragmatic "action tags" such as $\[propose\\\_motion\]$\. Using this pipeline, we release three public datasets of government deliberation \(Appellate Court hearings, School Board meetings, and Municipal Council sessions\) and fine\-tune LLM personas on this action\-aware data\. We evaluate simulations along four dimensions: persona fidelity, persona consistency, institutional fidelity, and behavioral coherence\. Action\-aware fine\-tuning cuts perplexity by 67%, doubles classifier\-based persona fidelity, increases vote attempts by up to $3\.6\\times$, and improves deliberative responsiveness by up to 70%\. Human evaluations show that simulated excerpts are often hard to distinguish from real deliberations, indicating a practical foundation for data\-grounded civic simulation studies\.

## Submission history

From: Scott Merrill \[[view email](https://arxiv.org/show-email/daa16231/2511.17813)\] **[\[v1\]](https://arxiv.org/abs/2511.17813v1)**Fri, 21 Nov 2025 22:07:33 UTC \(1,726 KB\) **[\[v2\]](https://arxiv.org/abs/2511.17813v2)**Sat, 14 Mar 2026 18:19:24 UTC \(1,615 KB\) **\[v3\]**Fri, 10 Jul 2026 02:38:19 UTC \(5,117 KB\)

Similar Articles

Low-Agreeableness Persona Conditioning for Safe LLM Fine-Tuning

arXiv cs.CL

This paper introduces a persona-driven rewriting pipeline that conditions LLM fine-tuning on low agreeableness to reduce jailbreak susceptibility and harmful outputs while preserving conversational warmth, without requiring safety labels or changes to training objective.