https://arxiv.org/abs/2501.05767
[2501.05767] Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large...
Abstract page for arXiv paper 2501.05767: Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large Language Models
https://migician-vg.github.io/
Migician
The First Free-Form Multi-Image Grounding MLLM.