Revolutionizing Dementia Care: Enhancing Talk Therapy with Fine‐Tuned Large Language Models Using GPT Self‐Generated Data
Jia Xu, Tianyi Wei, Bojian Hou, George Demiris, Li Shen · 2025
Read Revolutionizing Dementia Care: Enhancing Talk Therapy with Fine‐Tuned Large Language Models Using GPT Self‐Generated Data by Jia Xu, Tianyi Wei, Bojian Hou, George Demiris, Li Shen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11712389.txt.
Description
Jia Xu and four co-authors fine-tune chat models into dementia talk therapists in a 2025 paper. Talk therapy eases dementia's burden, but therapists are scarce. A general chatbot lacks the bedside manner and focus of a trained counselor, so the team built a specialist.
They had a commercial model write ten thousand practice question-and-answer pairs spanning depression, anxiety, sleep trouble, and more. Five open models were then tuned on that synthetic set with a light-weight method on a single graphics card. The tuned models answered test cases drawn from a counseling collection with a more professional tone and grip of complex feelings. The authors pitch the approach as a scalable stand-in where human care is thin.
Catalog record
| CREATOR | Jia Xu, Tianyi Wei, Bojian Hou, George Demiris, Li Shen |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 3 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |