Comparison of three large language models in postoperative rehabilitation question answering after anterior cruciate ligament reconstruction based on expert ratings
Tiange Xia, Yijie Zhang, Shaoshuo Li, Yi Zhou, Wenyu Tian, Zhiwei Jiang, Yang Shao, Jianwei Wang · 2026
Read Comparison of three large language models in postoperative rehabilitation question answering after anterior cruciate ligament reconstruction based on expert ratings by Tiange Xia, Yijie Zhang, Shaoshuo Li, Yi Zhou, Wenyu Tian, Zhiwei Jiang, Yang Shao, Jianwei Wang (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/pmc-13471468.txt.
Description
Xia and co-authors had orthopaedic experts grade three chatbots answering patient questions about rehabilitation after knee ligament surgery. Their 2026 blinded study sent 30 English questions, spanning five recovery phases after anterior cruciate ligament reconstruction, to GPT-5.4, Doubao, and MiniMax-M2.7 on a single day in March 2026.
Five clinicians scored all 90 responses on accuracy, safety, stage-fit, completeness, and understandability, without knowing which model wrote which answer. GPT-5.4 earned a higher overall score than Doubao, a gap rated highly significant after adjustment, while each model showed a distinct profile across the five domains. The authors warn that fluent wording is not enough for phase-sensitive rehab advice and that these benchmarks do not prove patient benefit.
Catalog record
| CREATOR | Tiange Xia, Yijie Zhang, Shaoshuo Li, Yi Zhou, Wenyu Tian, Zhiwei Jiang, Yang Shao, Jianwei Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |