WorldKV: Efficient World Memory with World Retrieval and Compression Paper • 2605.22718 • Published May 21 • 42
FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching Paper • 2605.20910 • Published May 20 • 29
Representation Alignment for Just Image Transformers is not Easier than You Think Paper • 2603.14366 • Published Mar 15 • 13