Hierarchical prompting with reasoning large language models for immune checkpoint inhibitor-associated myocarditis evidence extraction pilot study

Enshuo Hsu, Sheng-Chieh Lu, Sara Ebrahimi, Noah Beinart, Keila C Ostos-Mendoza, Anita Deswal, Kirk Roberts, Trey Kell, Jeff Jin, Nicolas L Palaskas · 2026

Read Hierarchical prompting with reasoning large language models for immune checkpoint inhibitor-associated myocarditis evidence extraction pilot study by Enshuo Hsu, Sheng-Chieh Lu, Sara Ebrahimi, Noah Beinart, Keila C Ostos-Mendoza, Anita Deswal, Kirk Roberts, Trey Kell, Jeff Jin, Nicolas L Palaskas (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-13602418.txt.

Description

Enshuo Hsu and co-authors pilot in 2026 a prompting scheme that pulls hard evidence out of cancer clinic notes. Immunotherapy can inflame the heart muscle, a rare but deadly side effect called myocarditis, and research on it needs data locked in free text. The team collected notes of confirmed cases at one center and set language models to extract structured facts.

Three techniques were tested in layers: model reasoning, context shaping, and hierarchical prompting that breaks the job into steps. A cardio-oncologist's hand-built record served as the truth to beat. Targets included tissue signs, dangerous rhythms, heart failure, and stroke.

Catalog record

CREATOREnshuo Hsu, Sheng-Chieh Lu, Sara Ebrahimi, Noah Beinart, Keila C Ostos-Mendoza, Anita Deswal, Kirk Roberts, Trey Kell, Jeff Jin, Nicolas L Palaskas
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13602418

Browse the full library · About the Tower of Babel