analiza aiqbench
This research demonstrates the feasibility of using large language models to create scalable, multimodal simulation environments for training psychotherapists in Cognitive Behavioural Therapy. The study highlights critical nuances in model performance, noting that while larger models improve diagnostic accuracy and speech-to-speech interactions yield better fidelity than text-only alternatives, current systems still tend to overestimate trainee competence without expert calibration. For the AI community, this work underscores the potential of GenAI as a tool for high-stakes professional education while flagging necessary safeguards against hallucinated feedback in sensitive therapeutic contexts.
MyMentorLLM: A psychotherapy GenAI environment with multimodal voice/text patients, trainees and experts for deliberate practice
arXiv:2607.25667v1 Announce Type: cross Abstract: Psychotherapists need repeated training and supervision by experts; however, scalability is problematic. Here we present MyMentorLLM, a multimodal voice- and text-based simulation environment for deliberate practice, used to generate 2,100 complete Cognitive Behavioural Therapy (CBT) training sessions. Each session links a DSM-5-TR-grounded patient (with major depressive, generalised anxiety or borderline personality disorder), a therapist-in-training and an expert supervisor. As an initial implementation, we adopted CBT because its structured procedures and competency-based supervision facilitate standardised simulation and evaluation. Sessions were analysed for emotional dynamics, therapeutic competence and diagnostic accuracy. Simulated patients expressed disorder-congruent emotional profiles, which trainee therapists mirrored as in real human counselling. The quality of supervision differed across LLMs: while most models overestimated trainees' competences, native speech-to-speech was closest to human scores. Supervisors' feedback led to better diagnoses in simulated psychotherapists in 5 out of 7 LLMs, and symptom identification accuracy increased with model size. This work shows that simulation of deliberate practice is possible for CBT training, although patient fidelity, calibration of supervisors, and harmful feedback should be evaluated together.
Abstract:Psychotherapists need repeated training and supervision by experts; however, scalability is problematic. Here we present MyMentorLLM, a multimodal voice- and text-based simulation environment for deliberate practice, used to generate 2,100 complete Cognitive Behavioural Therapy (CBT) training sessions. Each session links a DSM-5-TR-grounded patient (with major depressive, generalised anxiety or borderline personality disorder), a therapist-in-training and an expert supervisor. As an initial implementation, we adopted CBT because its structured procedures and competency-based supervision facilitate standardised simulation and evaluation. Sessions were analysed for emotional dynamics, therapeutic competence and diagnostic accuracy. Simulated patients expressed disorder-congruent emotional profiles, which trainee therapists mirrored as in real human counselling. The quality of supervision differed across LLMs: while most models overestimated trainees' competences, native speech-to-speech was closest to human scores. Supervisors' feedback led to better diagnoses in simulated psychotherapists in 5 out of 7 LLMs, and symptom identification accuracy increased with model size. This work shows that simulation of deliberate practice is possible for CBT training, although patient fidelity, calibration of supervisors, and harmful feedback should be evaluated together.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2607.25667 [cs.CL]
(or arXiv:2607.25667v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2607.25667 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 270 cuvinte
Subiecte conexe
34 Amazon Research Awards Build on Trainium recipients announced
Amazon Science · 5 aug. 2026
Amazon is investing in the Lean Focused Research Organization
Amazon Science · 26 iul. 2026
Build real-time voice applications with vLLM-Omni on SageMaker AI – Part 1
AWS · 28 sept. 2026
[AINews] Agents disprove Polysemanticity; greatly improve Interpretability & Alignment
Latent Space · 16 oct. 2026