Large Language Models for Distress Rating in Korean Psycho-Oncology Interviews: Exploratory Clinician-Benchmarked Evaluation Study
Jaehyun Kim, Kyung-Lak Son, Chan-Woo Yeom, Won-Hyoung Kim, Sun Hyung Lee, Joon Sung Shin, Hyunsun Yang, Daehun Yoo, Bong-Jin Hahm · 2026
Read Large Language Models for Distress Rating in Korean Psycho-Oncology Interviews: Exploratory Clinician-Benchmarked Evaluation Study by Jaehyun Kim, Kyung-Lak Son, Chan-Woo Yeom, Won-Hyoung Kim, Sun Hyung Lee, Joon Sung Shin, Hyunsun Yang, Daehun Yoo, Bong-Jin Hahm (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-13557317.txt.
Description
Jaehyun Kim and co-authors test whether large language models can rate distress from real interviews with Korean cancer patients, in a study from 2026. Distress is common in cancer care, but interview-based screening is hard to scale. Most past tests of such models used records written by clinicians or translated text, not real speech.
Just over one hundred adults in South Korea gave recorded interviews. Trained specialists in cancer mental health rated thirty-nine items live, with time stamps. Three leading models scored the same items from the Korean text and gave short reasons. The study compared each model's scores with the clinicians' marks. It also sorted the kinds of errors the models made, as judged by clinicians. Direction of bias was part of the check.
Catalog record
| CREATOR | Jaehyun Kim, Kyung-Lak Son, Chan-Woo Yeom, Won-Hyoung Kim, Sun Hyung Lee, Joon Sung Shin, Hyunsun Yang, Daehun Yoo, Bong-Jin Hahm |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |