Large Language Models Can Help Make Better Image Classifiers
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only! Grab it Discover how follow-up differential descriptions (FuDD) can enhance zero-shot image classification by tailoring class descriptions for each dataset in this 41-minute research talk. Explore how PhD Student Reza Esfandiarpoor from Brown University demonstrates FuDD's ability to identify ambiguous classes for each image and employ large language models (LLMs) to produce more distinguishing descriptions. Learn about FuDD's performance compared to few-shot adaptation methods, the challenges it excels at, and how to integrate it into your workflow. Gain insights into improving vision-language models through the cooperation of VLMs and LLMs, and access additional resources for further exploration of this cutting-edge AI research in image classification.