Diffusion-based or flow-based models have achieved significant progress in video synthesis but require multiple iterative sampling steps, which incurs substantial computational overhead. While many distillation methods that are solely based on trajectory-preserving or distribution-matching have been developed to accelerate video generation models, these approaches often suffer from performance breakdown or increased artifacts in few-step settings. To address these limitations, we propose SwiftVideo, a unified and stable distillation framework that combines the advantages of trajectory-preserving and distribution-matching strategies. Our approach introduces continuous-time consistency distillation to ensure precise preservation of ODE trajectories. Subsequently, We propose a dual-perspective alignment encompassing distribution alignment between synthetic and real data along with trajectory alignment across different inference steps. Our method maintains high-quality video generation while substantially reducing the number of inference steps. Quantitative evaluations on the OpenVid-1M benchmark demonstrate that our method significantly outperforms existing approaches in few-step video generation.

Knowledge Distillation (KD) is a key technique for compressing Large-scale Language Models (LLMs), but prevailing logit-based methods employ static strategies misaligned with the student’s dynamic learning process. By treating all tokens indiscriminately with a fixed temperature, these methods result in suboptimal knowledge transfer. To address this, we propose LLM-oriented token-Adaptive Knowledge Distillation (AdaKD), a framework that adapts the distillation process to each token’s real-time learning state. AdaKD consists of two synergistic modules driven by a unified token difficulty metric. First, the Loss-driven Adaptive Token Focusing (LATF) module dynamically concentrates distillation on valuable tokens by monitoring the student’s learning stability. Second, Inverse Difficulty Temperature Scaling (IDTS) introduces a counterintuitive token-level temperature: low for difficult tokens to target error correction, and high for easy tokens to learn the teacher’s smooth output distribution for better generalization. As a plug-and-play framework, AdaKD consistently improves performance across diverse distillation methods, model architectures, and benchmarks.

This work aims at a challenging task: human action reaction synthesis, i.e., generating human reactions conditioned on the action sequence of another person. Currently, autoregressive modeling approaches with vector quantization (VQ) have achieved remarkable performance in motion generation tasks. However, VQ has inherent disadvantages, including quantization information loss, low codebook utilization, etc. In addition, while dividing the body into separate units can be beneficial, the computational complexity needs to be considered. Also, the importance of mutual perception among units is often neglected. In this work, we propose MARRS, a novel framework designed to generate coordinated and fine-grained reaction motions us ing continuous representations. Initially, we present the Unit distinguished Motion Variational AutoEncoder (UD-VAE), which segments the entire body into distinct body and hand units, encoding each independently. Subsequently, we propose Action Conditioned Fusion (ACF), which involves randomly masking a subset of reactive tokens and extracting specific information about the body and hands from the active tokens. Furthermore, we introduce Mutual Unit Modulation (MUM) to facilitate interaction between body and hand units by using the information from one unit to adaptively modulate the other. Finally, for the diffusion model, we employ a compact MLP as a noise predictor for each distinct body unit and incorporate the diffusion loss to model the probability distribution of each token. Both quantitative and qualitative results demonstrate that our method achieves superior performance. The code will be released upon acceptance.

The performance of anomaly inspection in industrial manufacturing is constrained by the scarcity of anomaly data. To overcome this challenge, researchers have started employing anomaly generation approaches to augment the anomaly dataset. However, existing anomaly generation methods suffer from limited diversity in the generated anomalies and struggle to achieve a seamless blending of this anomaly with the original image. Moreover, the generated mask is usually not aligned with the generated anomaly. In this paper, we overcome these challenges from a new perspective, simultaneously generating a pair of the overall image and the corresponding anomaly part. We propose DualAnoDiff, a novel diffusion-based few-shot anomaly image generation model, which can generate diverse and realistic anomaly images by using a dual-interrelated diffusion model, where one of them is employed to generate the whole image while the other one generates the anomaly part. Moreover, we extract background and shape information to mitigate the distortion and blurriness phenomenon in few-shot image generation. Extensive experiments demonstrate the superiority of our proposed model over state-of-the-art methods in terms of diversity, realism and the accuracy of mask. Overall, our approach significantly improves the performance of downstream anomaly inspection tasks, including anomaly detection, anomaly localization, and anomaly classification tasks. Code will be made available.

Human-human motion generation is essential for understanding humans as social beings. Current methods fall into two main categories: single-person-based methods and separate modeling-based methods. To delve into this field, we abstract the overall generation process into a general framework MetaMotion, which consists of two phases: temporal modeling and interaction mixing. For temporal modeling, the single-person-based methods concatenate two people into a single one directly, while the separate modeling-based methods skip the modeling of interaction sequences. The inadequate modeling described above resulted in sub-optimal performance and redundant model parameters. In this paper, we introduce TIMotion (Temporal and Interactive Modeling), an efficient and effective framework for human-human motion generation. Specifically, we first propose Causal Interactive Injection to model two separate sequences as a causal sequence leveraging the temporal and causal properties. Then we present Role-Evolving Scanning to adjust to the change in the active and passive roles throughout the interaction. Finally, to generate smoother and more rational motion, we design Localized Pattern Amplification to capture short-term motion patterns. Extensive experiments on InterHuman and Inter-X demonstrate that our method achieves superior performance.

With the rapid development of diffusion models in image generation, the demand for more powerful and flexible controllable frameworks is increasing. Although existing methods can guide generation beyond text prompts, the challenge of effectively combining multiple conditional inputs while maintaining consistency with all of them remains unsolved. To address this, we introduce UniCombine, a DiT-based multi-conditional controllable generative framework capable of handling any combination of conditions, including but not limited to text prompts, spatial maps, and subject images. Specifically, we introduce a novel Conditional MMDiT Attention mechanism and incorporate a trainable LoRA module to build both the training-free and training-based versions. Additionally, we propose a new pipeline to construct SubjectSpatial200K, the first dataset designed for multi-conditional generative tasks covering both the subject-driven and spatially-aligned conditions. Extensive experimental results on multi-conditional generation demonstrate the outstanding universality and powerful capability of our approach with state-of-the-art performance.

Classifying an entire Whole Slide Image (WSI) in a single forward pass is challenging due to its vast resolution. Consequently, current effort on WSI classification resorts to multiple instance learning (MIL), using patch-wise instances to predict categories under image-wise supervision. However, recent MIL methods usually follow implicit instance selection strategy and ignore the effect from inherent patch category imbalances. In a statistical sense, negative patches dominate in WSIs and provide sufficient samples for accurate density estimation. Therefore, in this paper, we learn from anomaly detection and propose a deep MIL framework which learns a hybrid negative probability model to bootstrap discovery of potential positive lesion. We associate attention-based MIL approach with a regularization loss function to explicitly improve patch selection process in positive images. Experiments conducted on benchmarks of WSI recognition demonstrate that our method brings significant improvement to classic attention-based MIL baseline and achieves state-of-the-art performance.

Co-speech gesture generation is crucial for producing synchronized and realistic human gestures that accompany speech, enhancing the animation of lifelike avatars in virtual environments. While diffusion models have shown impressive capabilities, current approaches often overlook a wide range of modalities and their interactions, resulting in less dynamic and contextually varied gestures. To address these challenges, we present MambaGesture, a novel framework integrating a Mamba-based attention block, MambaAttn, with a multi-modality feature fusion module, SEAD. The MambaAttn block combines the sequential data processing strengths of the Mamba model with the contextual richness of attention mechanisms, enhancing the temporal coherence of generated gestures. SEAD adeptly fuses audio, text, style, and emotion modalities, employing disentanglement to deepen the fusion process and yield gestures with greater realism and diversity. Our approach, rigorously evaluated on the multi-modal BEAT dataset, demonstrates significant improvements in Fréchet Gesture Distance (FGD), diversity scores, and beat alignment, achieving state-of-the-art performance in co-speech gesture generation.

Recent advancements in the field of Diffusion Transformers have substantially improved the generation of high-quality 2D images, 3D videos, and 3D shapes. However, the effectiveness of the Transformer architecture in the domain of co-speech gesture generation remains relatively unexplored, as prior methodologies have predominantly employed the Convolutional Neural Network (CNNs) or simple a few transformer layers. In an attempt to bridge this research gap, we introduce a novel Masked Diffusion Transformer for co-speech gesture generation, referred to as MDT-A2G, which directly implements the denoising process on gesture sequences. To enhance the contextual reasoning capability of temporally aligned speech-driven gestures, we incorporate a novel Masked Diffusion Transformer. This model employs a mask modeling scheme specifically designed to strengthen temporal relation learning among sequence gestures, thereby expediting the learning process and leading to coherent and realistic motions. Apart from audio, Our MDT-A2G model also integrates multi-modal information, encompassing text, emotion, and identity. Furthermore, we propose an efficient inference strategy that diminishes the denoising computation by leveraging previously calculated results, thereby achieving a speedup with negligible performance degradation. Experimental results demonstrate that MDT-A2G excels in gesture generation, boasting a learning speed that is over 6× faster than traditional diffusion transformers and an inference speed that is 5.7× than the standard diffusion model. Our code is available at https://xiaofenmao.github.io/web-project/MDT-A2G/

Convolution-based and Transformer-based vision backbone networks process images into the grid or sequence structures, respectively, which are inflexible for capturing irregular objects. Though Vision GNN (ViG) adopts graph-level features for complex images, it has some issues, such as inaccurate neighbor node selection, expensive node information aggregation calculation, and over-smoothing in the deep layers. To address the above problems, we propose a Progressive Vision Graph (PVG) architecture for vision recognition task. Compared with previous works, PVG contains three main components: 1) Progressively Separated Graph Construction (PSGC) to introduce second-order similarity by gradually increasing the channel of the global graph branch and decreasing the channel of local branch as the layer deepens; 2) Neighbor nodes information aggregation and update module by using Max pooling and mathematical Expectation (MaxE) to aggregate rich neighbor information; 3) Graph error Linear Unit (GraphLU) to enhance low-value information in a relaxed form to reduce the compression of image detail information for alleviating the over-smoothing. Extensive experiments on mainstream benchmarks demonstrate the superiority of PVG over state-of-the-art methods, e.g., our PVG-S obtains 83.0% Top-1 accuracy on ImageNet-1K that surpasses GNN-based ViG-S by +0.9↑ with the parameters reduced by 18.5%, while the largest PVG-B obtains 84.2% that has +0.5↑ improvement than ViG-B. Furthermore, our PVG-S obtains +1.3↑ box AP and +0.4↑ mask AP gains than ViG-S on COCO dataset.