Vision-Language Grounding as Bidirectional Concept Correspondence Paper • 2608.07886 • Published 5 days ago • 5
Learning from Failures: Retrieval-Centric CoT via Hard Negatives for Unified Multimodal Retrieval Paper • 2608.06060 • Published 7 days ago • 40
AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities Paper • 2607.24821 • Published 27 days ago • 18
Motion Beyond Morphology: Bootstrapping Cross-Category Motion Transfer from Abstract Motion Representations Paper • 2608.01628 • Published 10 days ago • 22
Meshy T2: Fast Native Mesh Generation with Flow Matching Paper • 2607.28675 • Published 16 days ago • 56