@inproceedings{vu-etal-2026-roleplaying,
title = "Roleplaying with Structure: Synthetic Therapist-Client Conversation Generation from Questionnaires",
author = "Vu, Doan Nam Long and
Tan, Rui and
Francke, Svenja Jule and
Moench, Lena and
Woiwod, Daniel and
Thomas-Odenthal, Florian and
Stroth, Sanna and
Kircher, Tilo and
Hermann, Christiane and
Dannlowski, Udo and
Jamalabadi, Hamidreza and
Balloccu, Simone and
Ji, Shaoxiong",
editor = "Atwell, Katherine and
Borah, Angana and
Dementieva, Daryna and
Kreiss, Elisa and
Kotonya, Neema and
Liu, Jiarui and
Olson, Liz and
Wan, Ruyuan and
Zhao, Jieyu",
booktitle = "Proceedings of the Fifth Workshop on {NLP} for Positive Impact ({NLP}4{PI})",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.nlp4pi-1.33/",
pages = "395--417",
ISBN = "979-8-89176-495-8",
abstract = "Large Language Models (LLMs) are promising tools for synthetic data generation in mental health. However, privacy policies and restrictions forced previous work to rely mainly on generic information. We present a comprehensive corpus of synthetic therapist-client conversations generated through LLMs. We construct our generation pipeline, SQPsych (Structured Questionnaire-based Psychotherapy), which uses real structured client profiles and psychological questionnaires without leaking any sensitive data. We fine-tune various openweight LLMs on our generated corpus, SQPsychConv, and test them through both automatic benchmarks and human evaluation with trained psychotherapists. We find that standard benchmarks do not adequately capture the strengths of our dataset, but expert judgment shows that SQPsych makes LLMs significantly better at therapist roleplaying. Experts also consistently prefer therapy sessions generated by our models compared to other mental-health-oriented LLMs. We release our code, fine-tuned models SQPsychLLM, and corpora at https://ai-mh.github.io/SQPsych.html."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vu-etal-2026-roleplaying">
<titleInfo>
<title>Roleplaying with Structure: Synthetic Therapist-Client Conversation Generation from Questionnaires</title>
</titleInfo>
<name type="personal">
<namePart type="given">Doan</namePart>
<namePart type="given">Nam</namePart>
<namePart type="given">Long</namePart>
<namePart type="family">Vu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Svenja</namePart>
<namePart type="given">Jule</namePart>
<namePart type="family">Francke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lena</namePart>
<namePart type="family">Moench</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Woiwod</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Florian</namePart>
<namePart type="family">Thomas-Odenthal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sanna</namePart>
<namePart type="family">Stroth</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tilo</namePart>
<namePart type="family">Kircher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christiane</namePart>
<namePart type="family">Hermann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Udo</namePart>
<namePart type="family">Dannlowski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hamidreza</namePart>
<namePart type="family">Jamalabadi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simone</namePart>
<namePart type="family">Balloccu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shaoxiong</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifth Workshop on NLP for Positive Impact (NLP4PI)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Katherine</namePart>
<namePart type="family">Atwell</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Angana</namePart>
<namePart type="family">Borah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daryna</namePart>
<namePart type="family">Dementieva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elisa</namePart>
<namePart type="family">Kreiss</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Neema</namePart>
<namePart type="family">Kotonya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiarui</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liz</namePart>
<namePart type="family">Olson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruyuan</namePart>
<namePart type="family">Wan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jieyu</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-495-8</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) are promising tools for synthetic data generation in mental health. However, privacy policies and restrictions forced previous work to rely mainly on generic information. We present a comprehensive corpus of synthetic therapist-client conversations generated through LLMs. We construct our generation pipeline, SQPsych (Structured Questionnaire-based Psychotherapy), which uses real structured client profiles and psychological questionnaires without leaking any sensitive data. We fine-tune various openweight LLMs on our generated corpus, SQPsychConv, and test them through both automatic benchmarks and human evaluation with trained psychotherapists. We find that standard benchmarks do not adequately capture the strengths of our dataset, but expert judgment shows that SQPsych makes LLMs significantly better at therapist roleplaying. Experts also consistently prefer therapy sessions generated by our models compared to other mental-health-oriented LLMs. We release our code, fine-tuned models SQPsychLLM, and corpora at https://ai-mh.github.io/SQPsych.html.</abstract>
<identifier type="citekey">vu-etal-2026-roleplaying</identifier>
<location>
<url>https://aclanthology.org/2026.nlp4pi-1.33/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>395</start>
<end>417</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Roleplaying with Structure: Synthetic Therapist-Client Conversation Generation from Questionnaires
%A Vu, Doan Nam Long
%A Tan, Rui
%A Francke, Svenja Jule
%A Moench, Lena
%A Woiwod, Daniel
%A Thomas-Odenthal, Florian
%A Stroth, Sanna
%A Kircher, Tilo
%A Hermann, Christiane
%A Dannlowski, Udo
%A Jamalabadi, Hamidreza
%A Balloccu, Simone
%A Ji, Shaoxiong
%Y Atwell, Katherine
%Y Borah, Angana
%Y Dementieva, Daryna
%Y Kreiss, Elisa
%Y Kotonya, Neema
%Y Liu, Jiarui
%Y Olson, Liz
%Y Wan, Ruyuan
%Y Zhao, Jieyu
%S Proceedings of the Fifth Workshop on NLP for Positive Impact (NLP4PI)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-495-8
%F vu-etal-2026-roleplaying
%X Large Language Models (LLMs) are promising tools for synthetic data generation in mental health. However, privacy policies and restrictions forced previous work to rely mainly on generic information. We present a comprehensive corpus of synthetic therapist-client conversations generated through LLMs. We construct our generation pipeline, SQPsych (Structured Questionnaire-based Psychotherapy), which uses real structured client profiles and psychological questionnaires without leaking any sensitive data. We fine-tune various openweight LLMs on our generated corpus, SQPsychConv, and test them through both automatic benchmarks and human evaluation with trained psychotherapists. We find that standard benchmarks do not adequately capture the strengths of our dataset, but expert judgment shows that SQPsych makes LLMs significantly better at therapist roleplaying. Experts also consistently prefer therapy sessions generated by our models compared to other mental-health-oriented LLMs. We release our code, fine-tuned models SQPsychLLM, and corpora at https://ai-mh.github.io/SQPsych.html.
%U https://aclanthology.org/2026.nlp4pi-1.33/
%P 395-417
Markdown (Informal)
[Roleplaying with Structure: Synthetic Therapist-Client Conversation Generation from Questionnaires](https://aclanthology.org/2026.nlp4pi-1.33/) (Vu et al., NLP4PI 2026)
ACL
- Doan Nam Long Vu, Rui Tan, Svenja Jule Francke, Lena Moench, Daniel Woiwod, Florian Thomas-Odenthal, Sanna Stroth, Tilo Kircher, Christiane Hermann, Udo Dannlowski, Hamidreza Jamalabadi, Simone Balloccu, and Shaoxiong Ji. 2026. Roleplaying with Structure: Synthetic Therapist-Client Conversation Generation from Questionnaires. In Proceedings of the Fifth Workshop on NLP for Positive Impact (NLP4PI), pages 395–417, Budapest, Hungary. Association for Computational Linguistics.