MentalBART is part of the MentaLLaMA project, the first open-source large language model (LLM) series for interpretable mental health analysis with instruction-following capability. This model is finetuned based on the facebook/bart-large foundation model and the full IMHI instruction tuning data. The model is expected to make complex mental health analysis for various mental health conditions and give reliable explanations for each of its predictions. It is fine-tuned on the IMHI dataset with 75K high-quality natural language instructions to boost its performance in downstream tasks. We perform a comprehensive evaluation on the IMHI benchmark with 20K test samples. The result shows that MentalBART can achieve good performance in correctness and generates explanations. Although experiments on MentalBART show promising performance on interpretable mental health analysis, we stress that all predicted results and generated explanations should only used for non-clinical research, and the help-seeker should get assistance from professional psychiatrists or clinical practitioners. In addition, recent studies have indicated LLMs may introduce some potential bias, such as gender gaps.…
Модальности:
Генерация текста
Задача: Генерация текста
Автор: Tianlin668
Теги: bart, text2text-generation, mentalhealth, BART, text-generation-inference, en, endpoints_compatible
Лайков: 3 | Загрузок: 31
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.