COCONut-PanCap: Joint Panoptic Segmentation and Grounded Captions for Fine-Grained Understanding and Generation Paper β’ 2502.02589 β’ Published 8 days ago β’ 8
Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens Paper β’ 2501.07730 β’ Published 30 days ago β’ 16
Randomized Autoregressive Visual Generation Paper β’ 2411.00776 β’ Published Nov 1, 2024 β’ 17 β’ 3