Risk-centered benchmarking of large language models for AI-enabled counseling in chronic autoimmune thyroid eye disease
Fangqin Fei, Lu Xie, Jing Rao, Ziqi Liang, Juan Yang, Yunyun Zou, Ligang Jiang · 2026
Read Risk-centered benchmarking of large language models for AI-enabled counseling in chronic autoimmune thyroid eye disease by Fangqin Fei, Lu Xie, Jing Rao, Ziqi Liang, Juan Yang, Yunyun Zou, Ligang Jiang (2026) 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/PMC13311113.txt.
Description
Fangqin Fei and co-authors grade chatbots for thyroid eye disease advice in a 2026 benchmark study. Patients with the chronic autoimmune eye disease increasingly ask large language models for guidance, yet quality and safety proof was thin. A bank of 35 Chinese questions from guidelines and common patient asks was posed to five chatbots.
Two blinded experts scored replies on accuracy, logic, flow, safety, and plain access against a guideline standard. Reply speed and length varied widely by model. Scores split clearly on accuracy, logic, flow, and access, but not on safety. Longer replies did not mean better replies. Risk centered grading may guide patient use.
Catalog record
| CREATOR | Fangqin Fei, Lu Xie, Jing Rao, Ziqi Liang, Juan Yang, Yunyun Zou, Ligang Jiang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55132 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |