Related papers: REPA Works Until It Doesn't: Early-Stopped, Holist…

Semantic-rich features from Vision Foundation Models (VFMs) have been leveraged to enhance Latent Diffusion Models (LDMs). However, raw VFM features are typically high-dimensional and redundant, increasing the difficulty of learning and…

Distilling vision-language models into faster hybrid architectures, such as 3:1 Mamba-2/attention mixes, is now standard practice for making inference efficient. Aggregate benchmarks suggest that this works but they hide selective failures.…

Diffusion Transformers (DiTs) have achieved state-of-the-art performance in generative modeling, yet their high computational cost hinders real-time deployment. While feature caching offers a promising training-free acceleration solution by…

Test-time adaptation (TTA) addresses the unforeseen distribution shifts occurring during test time. In TTA, performance, memory consumption, and time consumption are crucial considerations. A recent diffusion-based TTA approach for…

Diffusion models (DMs) are a powerful generative framework that have attracted significant attention in recent years. However, the high computational cost of training DMs limits their practical applications. In this paper, we start with a…

Existing person re-identification (Re-ID) methods principally deploy the ImageNet-1K dataset for model initialization, which inevitably results in sub-optimal situations due to the large domain gap. One of the key challenges is that…

Diffusion models have significantly advanced generative AI, but they encounter difficulties when generating complex combinations of multiple objects. As the final result heavily depends on the initial seed, accurately ensuring the desired…

Scene rearrangement, like table tidying, is a challenging task in robotic manipulation due to the complexity of predicting diverse object arrangements. Web-scale trained generative models such as Stable Diffusion can aid by generating…

Diffusion models (DMs) produce high-quality images, yet their sampling remains costly when adapted to new domains. Distilled DMs are faster but typically remain confined within their teacher's domain. Thus, fast and high-quality generation…

Diffusion Transformers (DiTs) with billions of model parameters form the backbone of popular image and video generation models like DALL.E, Stable-Diffusion and SORA. Though these models are necessary in many low-latency applications like…

Vision transformers (ViTs) have recently obtained success in many applications, but their intensive computation and heavy memory usage at both training and inference time limit their generalization. Previous compression algorithms usually…

Diffusion transformers (DiTs) struggle to generate images at resolutions higher than their training resolutions. The primary obstacle is that the explicit positional encodings(PE), such as RoPE, need extrapolating to unseen positions which…

Reducing token count is crucial for efficient training and inference of latent diffusion models, especially at high resolution. A common strategy is to build high-compression image tokenizers with more channels per token. However, when…

Recent developments in large-scale pre-trained text-to-image diffusion models have significantly improved the generation of high-fidelity images, particularly with the emergence of diffusion transformer models (DiTs). Among diffusion…

This paper aims to accelerate video stream processing, such as object detection and semantic segmentation, by leveraging the temporal redundancies that exist between video frames. Instead of propagating and warping features using motion…

Despite promising progress in face swapping task, realistic swapped images remain elusive, often marred by artifacts, particularly in scenarios involving high pose variation, color differences, and occlusion. To address these issues, we…

Diffusion Transformers (DiTs) have proven effective in generating high-quality videos but are hindered by high computational costs. Existing video DiT sampling acceleration methods often rely on costly fine-tuning or exhibit limited…

This paper presents SPIE: a novel approach for semantic and structural post-training of instruction-based image editing diffusion models, addressing key challenges in alignment with user prompts and consistency with input images. We…

Video diffusion models have recently enabled high-quality video generation with ViT-based architectures, but remain computationally intensive because generation requires attention computation over long spatiotemporal sequences. Token…

Vision-Language-Action (VLA) models pre-trained on large, diverse datasets show remarkable potential for general-purpose robotic manipulation. However, a primary bottleneck remains in adapting these models to downstream tasks, especially…