CBT-Bench: Evaluating Large Language Models on Assisting Cognitive Behavior Therapy
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910800709419008 |
|---|---|
| author | Zhang, Mian Yang, Xianjun Zhang, Xinlu Labrum, Travis Chiu, Jamie C. Eack, Shaun M. Fang, Fei Wang, William Yang Chen, Zhiyu Zoey |
| author_facet | Zhang, Mian Yang, Xianjun Zhang, Xinlu Labrum, Travis Chiu, Jamie C. Eack, Shaun M. Fang, Fei Wang, William Yang Chen, Zhiyu Zoey |
| contents | There is a significant gap between patient needs and available mental health support today. In this paper, we aim to thoroughly examine the potential of using Large Language Models (LLMs) to assist professional psychotherapy. To this end, we propose a new benchmark, CBT-BENCH, for the systematic evaluation of cognitive behavioral therapy (CBT) assistance. We include three levels of tasks in CBT-BENCH: I: Basic CBT knowledge acquisition, with the task of multiple-choice questions; II: Cognitive model understanding, with the tasks of cognitive distortion classification, primary core belief classification, and fine-grained core belief classification; III: Therapeutic response generation, with the task of generating responses to patient speech in CBT therapy sessions. These tasks encompass key aspects of CBT that could potentially be enhanced through AI assistance, while also outlining a hierarchy of capability requirements, ranging from basic knowledge recitation to engaging in real therapeutic conversations. We evaluated representative LLMs on our benchmark. Experimental results indicate that while LLMs perform well in reciting CBT knowledge, they fall short in complex real-world scenarios requiring deep analysis of patients' cognitive structures and generating effective responses, suggesting potential future work. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_13218 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CBT-Bench: Evaluating Large Language Models on Assisting Cognitive Behavior Therapy Zhang, Mian Yang, Xianjun Zhang, Xinlu Labrum, Travis Chiu, Jamie C. Eack, Shaun M. Fang, Fei Wang, William Yang Chen, Zhiyu Zoey Computation and Language Artificial Intelligence Computers and Society There is a significant gap between patient needs and available mental health support today. In this paper, we aim to thoroughly examine the potential of using Large Language Models (LLMs) to assist professional psychotherapy. To this end, we propose a new benchmark, CBT-BENCH, for the systematic evaluation of cognitive behavioral therapy (CBT) assistance. We include three levels of tasks in CBT-BENCH: I: Basic CBT knowledge acquisition, with the task of multiple-choice questions; II: Cognitive model understanding, with the tasks of cognitive distortion classification, primary core belief classification, and fine-grained core belief classification; III: Therapeutic response generation, with the task of generating responses to patient speech in CBT therapy sessions. These tasks encompass key aspects of CBT that could potentially be enhanced through AI assistance, while also outlining a hierarchy of capability requirements, ranging from basic knowledge recitation to engaging in real therapeutic conversations. We evaluated representative LLMs on our benchmark. Experimental results indicate that while LLMs perform well in reciting CBT knowledge, they fall short in complex real-world scenarios requiring deep analysis of patients' cognitive structures and generating effective responses, suggesting potential future work. |
| title | CBT-Bench: Evaluating Large Language Models on Assisting Cognitive Behavior Therapy |
| topic | Computation and Language Artificial Intelligence Computers and Society |
| url | https://arxiv.org/abs/2410.13218 |