Performance, Failures, and Oversight of a Large Language Model Agent for Clinical Data Analysis: Evaluation Study
Yilan Wu, Dun Jack Fu, Yukun Zhou, Siegfried K Wagner, Pearse A Keane, Ivan Steenstra · 2026
Read Performance, Failures, and Oversight of a Large Language Model Agent for Clinical Data Analysis: Evaluation Study by Yilan Wu, Dun Jack Fu, Yukun Zhou, Siegfried K Wagner, Pearse A Keane, Ivan Steenstra (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-13552643.txt.
Description
Wu and co-authors evaluated a language model agent for clinical data work in a 2026 study. Such agents can write and run code from plain requests, which may widen access to data skills. The team tested which stages of a study the agent does well and where experts must watch.
Tests used a published eye study data set with three prompt levels and three ways of working, for twenty-seven runs in total. The agent made sound research questions and study plans, and its survival curves matched well. Code errors, though, repeated in the same way across runs, and some summaries held real clinical errors. Expert checks remain vital, the authors say.
Catalog record
| CREATOR | Yilan Wu, Dun Jack Fu, Yukun Zhou, Siegfried K Wagner, Pearse A Keane, Ivan Steenstra |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |