Exploiting large language models for diagnosing autism associated language disorders and identifying distinct features
Chuanbo Hu, Wenqi Li, Mindi Ruan, Xiangxu Yu, Shalaka Deshpande, Lynn K Paul, Shuo Wang, Xin Li · 2025
Read Exploiting large language models for diagnosing autism associated language disorders and identifying distinct features by Chuanbo Hu, Wenqi Li, Mindi Ruan, Xiangxu Yu, Shalaka Deshpande, Lynn K Paul, Shuo Wang, Xin Li (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-12708842.txt.
Description
Chuanbo Hu and co-authors use large language models to spot language traits tied to autism in a 2025 paper. Standard checks for such disorders lean on human judgment, which varies and takes time. The team asked whether a model could read speech or text and flag the patterns that mark the condition.
With no task specific training, their method beat baseline models by over ten points on sensitivity and on the share of positive calls that were right. Ten key features stood out, among them echoing heard words, swapping pronouns, and atypical word use. The authors offer the tool as a support for clinicians, not a stand in, and a base for tailored therapy plans.
Catalog record
| CREATOR | Chuanbo Hu, Wenqi Li, Mindi Ruan, Xiangxu Yu, Shalaka Deshpande, Lynn K Paul, Shuo Wang, Xin Li |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |