CBT-Bench: Evaluating Large Language Models on Assisting Cognitive Behavior Therapy
This addresses the gap in mental health support by assessing AI's potential to assist professional psychotherapy, though it is incremental as it focuses on evaluation rather than novel therapeutic methods.
The paper tackles the problem of evaluating large language models (LLMs) for assisting cognitive behavioral therapy (CBT) by proposing a new benchmark, CBT-BENCH, which includes tasks from basic knowledge to therapeutic response generation, and finds that LLMs perform well on basic tasks but fall short in complex, real-world scenarios.
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.