DICE: Discrete Inversion Enabling Controllable Editing for Multinomial Diffusion and Masked Generative Models Paper • 2410.08207 • Published Oct 10 • 18
Probabilistic Conceptual Explainers: Trustworthy Conceptual Explanations for Vision Foundation Models Paper • 2406.12649 • Published Jun 18 • 15
Multimodal Needle in a Haystack: Benchmarking Long-Context Capability of Multimodal Large Language Models Paper • 2406.11230 • Published Jun 17 • 34