Using deep vision-language models improves multi-task performance in assistance applications for endoscopic ENT surgery.

Richard Bieck Martin Sorge Katharina Heuermann Viktor Kunz Markus Pirlich Thomas Neumuth Using deep vision-language models improves multi-task performance in assistance applications for endoscopic ENT surgery. 2026 April 711-722 21 Int. J. Comput. Assist. Radiol. Surg. 4 https://doi.org/10.1007/s11548-025-03512-z db/journals/cars/cars21.html#BieckSHKPN26 streams/journals/cars