PaliGemma 2
Google DeepMind — Gemma
rich detail 21 method fields published
Evaluation
transfer of PaliGemma to MS COCO object detection and DocLayNet document layout detection using a pix2seq-inspired sequence augmentation approach ('detect all classes' prefix, noise boxes for confidence representation); mean average precision (mAP)
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.
rich detail 21 method fields published