FindIt benchmark for multimodal LLMs on visual detection

Arxiv CS.CV (computer vision)

FindIt is a format-informed visual detection benchmark for generalist multimodal LLMs. It evaluates models on structured tasks like object detection and layout analysis.

multimodalllmbenchmarks

Arxiv CS.CV (computer vision)·Eshika KhandelwalJingjing PanMingfang ZhangQuan KongLorenzo GarattoniHilde Kuehne

7 days ago