dorsal/arxiv
View SchemaHonesty-Aware Multi-Agent Framework for High-Fidelity Synthetic Data Generation in Digital Psychiatric Intake Doctor-Patient Interactions
| Authors | Xinyuan Zhang, Zijian Wang, Chang Dao, Juexiao Zhou |
|---|---|
| Categories | |
| ArXiv ID | 2601.09216vv1 |
| URL | https://arxiv.org/abs/2601.09216 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Data scarcity and unreliable self-reporting -- such as concealment or exaggeration -- pose fundamental challenges to psychiatric intake and assessment. We propose a multi-agent synthesis framework that explicitly models patient deception to generate high-fidelity, publicly releasable synthetic psychiatric intake records. Starting from DAIC-WOZ interviews, we construct enriched patient profiles and simulate a four-role workflow: a \emph{Patient} completes self-rated scales and participates in a semi-structured interview under a topic-dependent honesty state; an \emph{Assessor} selects instruments based on demographics and chief complaints; an \emph{Evaluator} conducts the interview grounded in rater-administered scales, tracks suspicion, and completes ratings; and a \emph{Diagnostician} integrates all evidence into a diagnostic summary. Each case links the patient profile, self-rated and rater-administered responses, interview transcript, diagnostic summary, and honesty state. We validate the framework through four complementary evaluations: diagnostic consistency and severity grading, chain-of-thought ablations, human evaluation of clinical realism and dishonesty modeling, and LLM-based comparative evaluation. The resulting corpus spans multiple disorders and severity levels, enabling controlled study of dishonesty-aware psychiatric assessment and the training and evaluation of adaptive dialogue agents.
{
"annotation_id": "3e108654-a2a6-48c8-8995-fae85ae92298",
"date_created": "2026-02-17T05:53:20.069000Z",
"date_modified": "2026-02-17T05:53:20.069000Z",
"file_hash": "762c0eb1abeb3de5d06ef33df9e2a886a35dcd17a63d95a6614142198e0f1794",
"private": false,
"record": {
"abstract": "Data scarcity and unreliable self-reporting -- such as concealment or exaggeration -- pose fundamental challenges to psychiatric intake and assessment. We propose a multi-agent synthesis framework that explicitly models patient deception to generate high-fidelity, publicly releasable synthetic psychiatric intake records. Starting from DAIC-WOZ interviews, we construct enriched patient profiles and simulate a four-role workflow: a \\emph{Patient} completes self-rated scales and participates in a semi-structured interview under a topic-dependent honesty state; an \\emph{Assessor} selects instruments based on demographics and chief complaints; an \\emph{Evaluator} conducts the interview grounded in rater-administered scales, tracks suspicion, and completes ratings; and a \\emph{Diagnostician} integrates all evidence into a diagnostic summary. Each case links the patient profile, self-rated and rater-administered responses, interview transcript, diagnostic summary, and honesty state. We validate the framework through four complementary evaluations: diagnostic consistency and severity grading, chain-of-thought ablations, human evaluation of clinical realism and dishonesty modeling, and LLM-based comparative evaluation. The resulting corpus spans multiple disorders and severity levels, enabling controlled study of dishonesty-aware psychiatric assessment and the training and evaluation of adaptive dialogue agents.",
"arxiv_id": "2601.09216",
"authors": [
"Xinyuan Zhang",
"Zijian Wang",
"Chang Dao",
"Juexiao Zhou"
],
"categories": [
"cs.DB"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Honesty-Aware Multi-Agent Framework for High-Fidelity Synthetic Data Generation in Digital Psychiatric Intake Doctor-Patient Interactions",
"url": "https://arxiv.org/abs/2601.09216",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "b8aab742-6c11-4709-843e-8b9d8e781b93",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}