🚀 Introducing Zooming without Zooming v2 (ZwZ2): learning to connect visual evidence across regions.
Trained on just 20.4K single-image examples generated by EvidenceConnect, ZwZ2 generalizes to multi-image long-context understanding, embodied spatial reasoning, and multimodal
All things open source at Ant Group. We aim to bring high caliber infrastructure FinTech OSS to the community.



