Optimizing a Multimodal Large Language Model for Ultrasound-Based Thyroid Nodule Malignancy Classification: A Comparative Study of Few-Shot Learning, Prompt Engineering, and Fine-Tuning

We also noted that the hybrid strategy significantly improved AUC over fine-tuning alone in the AUS cohort and TN5000 test split dataset, whereas the smaller improvement observed in DDTI did not reach statistical significance.

also in 111,027 other papers