Skip to content

Dataset · Psychotherapy-LLM

Cognitive behavioral therapy benchmark for LLMs

Psychotherapy-LLM/CBT-Bench

English dataset for evaluating language model skills in cognitive behavioral therapy (CBT).

What's inside

The dataset is an English text benchmark with a volume of 1,000 to 10,000 records. The tasks are divided into three difficulty levels: checking basic CBT knowledge, understanding the cognitive model, and generating therapeutic responses. The data is designed to evaluate model capabilities in supporting various stages of psychological assistance.

What you can do with it

  • Evaluating and comparing LLMs in psychological support and CBT tasks
  • Validating specialized medical and therapeutic assistants
  • Analyzing model ability to follow cognitive behavioral therapy protocols
Tasks
Question answering, Text generation, Text classification
Languages
English
Size
1K–10K rows
Downloads this month
3.4K
Likes
46

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Psychotherapy-LLM/CBT-Bench")