12 papers · 1 filter
Focusing on What Matters: Saliency-Harnessing Accurate Routing for Diffusion MoE
Haoyou Deng, Keyu Yan, Chaojie Mao +4
Mixture-of-Experts (MoE) architectures have emerged as a powerful paradigm for scaling diffusion models in visual generation. Recent advancements have focused on adaptively allocat…
DenseGRPO: From Sparse to Dense Reward for Flow Matching Model Alignment
Haoyou Deng, Keyu Yan, Chaojie Mao +4
Recent GRPO-based approaches built on flow matching models have shown remarkable improvements in human preference alignment for text-to-image generation. Nevertheless, they still s…
Unison: A Fully Automatic, Task-Universal, and Low-Cost Framework for Unified Understanding and Generation
Shihao Zhao, Yitong Chen, Zeyinzi Jiang +5
Unified understanding and generation is a highly appealing research direction in multimodal learning. There exist two approaches: one trains a transformer via an auto-regressive pa…
Lego-Edit: A General Image Editing Framework with Model-Level Bricks and MLLM Builder
Qifei Jia, Yu Liu, Yajie Chai +6
Instruction-based image editing has garnered significant attention due to its direct interaction with users. However, real-world user instructions are immensely diverse, and existi…
AdPO: Enhancing the Adversarial Robustness of Large Vision-Language Models with Preference Optimization
Chaohu Liu, Tianyi Gui, Yu Liu +1
Large Vision-Language Models (LVLMs), such as GPT-4o and LLaVA, have recently witnessed remarkable advancements and are increasingly being deployed in real-world applications. Howe…
Rethinking Video Tokenization: A Conditioned Diffusion-based Approach
Nianzu Yang, Pandeng Li, Liming Zhao +8
Existing video tokenizers typically use the traditional Variational Autoencoder (VAE) architecture for video compression and reconstruction. However, to achieve good performance, i…