Zhen Xing

Zhen Xing, Qijun Feng, Haoran Chen et al.

The recent wave of AI-generated content (AIGC) has witnessed substantial success in computer vision, with the diffusion model playing a crucial role in this achievement. Due to their impressive generative capabilities, diffusion models are gradually superseding methods based on GANs and auto-regressive Transformers, demonstrating exceptional performance not only in image generation and editing, but also in the realm of video-related research. However, existing surveys mainly focus on diffusion models in the context of image generation, with few up-to-date reviews on their application in the video domain. To address this gap, this paper presents a comprehensive review of video diffusion models in the AIGC era. Specifically, we begin with a concise introduction to the fundamentals and evolution of diffusion models. Subsequently, we present an overview of research on diffusion models in the video domain, categorizing the work into three key areas: video generation, video editing, and other video understanding tasks. We conduct a thorough review of the literature in these three key areas, including further categorization and practical contributions in the field. Finally, we discuss the challenges faced by research in this domain and outline potential future developmental trends. A comprehensive list of video diffusion models studied in this survey is available at https://github.com/ChenHsing/Awesome-Video-Diffusion-Models.

Zejia Weng, Xitong Yang, Zhen Xing et al.

Video diffusion models are able to generate high-quality videos by learning strong spatial-temporal priors on large-scale datasets. In this paper, we aim to investigate whether such priors derived from a generative process are suitable for video recognition, and eventually joint optimization of generation and recognition. Building upon Stable Video Diffusion, we introduce GenRec, the first unified framework trained with a random-frame conditioning process so as to learn generalized spatial-temporal representations. The resulting framework can naturally supports generation and recognition, and more importantly is robust even when visual inputs contain limited information. Extensive experiments demonstrate the efficacy of GenRec for both recognition and generation. In particular, GenRec achieves competitive recognition performance, offering 75.8% and 87.2% accuracy on SSV2 and K400, respectively. GenRec also performs the best on class-conditioned image-to-video generation, achieving 46.5 and 49.3 FVD scores on SSV2 and EK-100 datasets. Furthermore, GenRec demonstrates extraordinary robustness in scenarios that only limited frames can be observed. Code will be available at https://github.com/wengzejia1/GenRec.

Zhen Xing, Qi Dai, Han Hu et al.

Semi-supervised action recognition is a challenging but critical task due to the high cost of video annotations. Existing approaches mainly use convolutional neural networks, yet current revolutionary vision transformer models have been less explored. In this paper, we investigate the use of transformer models under the SSL setting for action recognition. To this end, we introduce SVFormer, which adopts a steady pseudo-labeling framework (ie, EMA-Teacher) to cope with unlabeled video samples. While a wide range of data augmentations have been shown effective for semi-supervised image classification, they generally produce limited results for video recognition. We therefore introduce a novel augmentation strategy, Tube TokenMix, tailored for video data where video clips are mixed via a mask with consistent masked tokens over the temporal axis. In addition, we propose a temporal warping augmentation to cover the complex temporal variation in videos, which stretches selected frames to various temporal durations in the clip. Extensive experiments on three datasets Kinetics-400, UCF-101, and HMDB-51 verify the advantage of SVFormer. In particular, SVFormer outperforms the state-of-the-art by 31.5% with fewer training epochs under the 1% labeling rate of Kinetics-400. Our method can hopefully serve as a strong benchmark and encourage future search on semi-supervised action recognition with Transformer networks.

Zhen Xing, Qi Dai, Han Hu et al.

The recent wave of AI-generated content has witnessed the great development and success of Text-to-Image (T2I) technologies. By contrast, Text-to-Video (T2V) still falls short of expectations though attracting increasing interests. Existing works either train from scratch or adapt large T2I model to videos, both of which are computation and resource expensive. In this work, we propose a Simple Diffusion Adapter (SimDA) that fine-tunes only 24M out of 1.1B parameters of a strong T2I model, adapting it to video generation in a parameter-efficient way. In particular, we turn the T2I model for T2V by designing light-weight spatial and temporal adapters for transfer learning. Besides, we change the original spatial attention to the proposed Latent-Shift Attention (LSA) for temporal consistency. With similar model architecture, we further train a video super-resolution model to generate high-definition (1024x1024) videos. In addition to T2V generation in the wild, SimDA could also be utilized in one-shot video editing with only 2 minutes tuning. Doing so, our method could minimize the training effort with extremely few tunable parameters for model adaptation.

Zhen Xing, Hengduo Li, Zuxuan Wu et al.

The performance of existing single-view 3D reconstruction methods heavily relies on large-scale 3D annotations. However, such annotations are tedious and expensive to collect. Semi-supervised learning serves as an alternative way to mitigate the need for manual labels, but remains unexplored in 3D reconstruction. Inspired by the recent success of semi-supervised image classification tasks, we propose SSP3D, a semi-supervised framework for 3D reconstruction. In particular, we introduce an attention-guided prototype shape prior module for guiding realistic object reconstruction. We further introduce a discriminator-guided module to incentivize better shape generation, as well as a regularizer to tolerate noisy training samples. On the ShapeNet benchmark, the proposed approach outperforms previous supervised methods by clear margins under various labeling ratios, (i.e., 1%, 5% , 10% and 20%). Moreover, our approach also performs well when transferring to real-world Pix3D datasets under labeling ratios of 10%. We also demonstrate our method could transfer to novel categories with few novel supervised data. Experiments on the popular ShapeNet dataset show that our method outperforms the zero-shot baseline by over 12% and we also perform rigorous ablations and analysis to validate our approach.

Zhen Xing, Qi Dai, Zihao Zhang et al.

Diffusion models have achieved significant success in image and video generation. This motivates a growing interest in video editing tasks, where videos are edited according to provided text descriptions. However, most existing approaches only focus on video editing for short clips and rely on time-consuming tuning or inference. We are the first to propose Video Instruction Diffusion (VIDiff), a unified foundation model designed for a wide range of video tasks. These tasks encompass both understanding tasks (such as language-guided video object segmentation) and generative tasks (video editing and enhancement). Our model can edit and translate the desired results within seconds based on user instructions. Moreover, we design an iterative auto-regressive method to ensure consistency in editing and enhancing long videos. We provide convincing generative results for diverse input videos and written instructions, both qualitatively and quantitatively. More examples can be found at our website https://ChenHsing.github.io/VIDiff.

Hui Zhang, Zuxuan Wu, Zhen Xing et al.

Diffusion models, as a type of generative model, have achieved impressive results in generating images and videos conditioned on textual conditions. However, the generation process of diffusion models involves denoising dozens of steps to produce photorealistic images/videos, which is computationally expensive. Unlike previous methods that design ``one-size-fits-all'' approaches for speed up, we argue denoising steps should be sample-specific conditioned on the richness of input texts. To this end, we introduce AdaDiff, a lightweight framework designed to learn instance-specific step usage policies, which are then used by the diffusion model for generation. AdaDiff is optimized using a policy gradient method to maximize a carefully designed reward function, balancing inference time and generation quality. We conduct experiments on three image generation and two video generation benchmarks and demonstrate that our approach achieves similar visual quality compared to the baseline using a fixed 50 denoising steps while reducing inference time by at least 33%, going as high as 40%. Furthermore, our method can be used on top of other acceleration methods to provide further speed benefits. Lastly, qualitative analysis shows that AdaDiff allocates more steps to more informative prompts and fewer steps to simpler prompts.

Yujie Wei, Yujin Han, Zhekai Chen et al.

Video generation is rapidly evolving from single-shot synthesis to complex multi-shot audio-video (MSAV) narratives to meet real-world demands. However, evaluating such frontier models remains a fundamental challenge. Existing benchmarks are limited in scope and data diversity, and rely on rigid evaluation pipelines, preventing systematic and reliable assessment of modern MSAV models. To bridge these gaps, we introduce MSAVBench, the first comprehensive benchmark and adaptive hybrid evaluation framework for multi-shot audio-video generation. Our benchmark spans four key dimensions, video, audio, shot, and reference, covering diverse task settings, varying shot counts of up to 15, and challenging non-realistic scenarios. Our evaluation framework improves robustness through an adaptive self-correction mechanism for shot segmentation, instance-wise rubrics for subjective metrics, and tool-grounded evidence extraction for complex judgments. Furthermore, MSAVBench achieves high alignment with human judgments, reaching a Spearman rank correlation of 91.5%. Our systematic evaluation of 19 state-of-the-art closed- and open-source models shows that current systems still struggle with director-level control and fine-grained audio-visual synchronization, while modular or agentic generation pipelines offer a promising path toward narrowing the gap between open- and closed-source models. We will release the benchmark data and evaluation code to facilitate future research.

Zhixin Ling, Zhen Xing, Xiangdong Zhou et al.

In panorama understanding, the widely used equirectangular projection (ERP) entails boundary discontinuity and spatial distortion. It severely deteriorates the conventional CNNs and vision Transformers on panoramas. In this paper, we propose a simple yet effective architecture named PanoSwin to learn panorama representations with ERP. To deal with the challenges brought by equirectangular projection, we explore a pano-style shift windowing scheme and novel pitch attention to address the boundary discontinuity and the spatial distortion, respectively. Besides, based on spherical distance and Cartesian coordinates, we adapt absolute positional embeddings and relative positional biases for panoramas to enhance panoramic geometry information. Realizing that planar image understanding might share some common knowledge with panorama understanding, we devise a novel two-stage learning framework to facilitate knowledge transfer from the planar images to panoramas. We conduct experiments against the state-of-the-art on various panoramic tasks, i.e., panoramic object detection, panoramic classification, and panoramic layout estimation. The experimental results demonstrate the effectiveness of PanoSwin in panorama understanding.

Zhen Xing, Yijiang Chen, Zhixin Ling et al.

3D reconstruction of novel categories based on few-shot learning is appealing in real-world applications and attracts increasing research interests. Previous approaches mainly focus on how to design shape prior models for different categories. Their performance on unseen categories is not very competitive. In this paper, we present a Memory Prior Contrastive Network (MPCN) that can store shape prior knowledge in a few-shot learning based 3D reconstruction framework. With the shape memory, a multi-head attention module is proposed to capture different parts of a candidate shape prior and fuse these parts together to guide 3D reconstruction of novel categories. Besides, we introduce a 3D-aware contrastive learning method, which can not only complement the retrieval accuracy of memory network, but also better organize image features for downstream tasks. Compared with previous few-shot 3D reconstruction methods, MPCN can handle the inter-class variability without category annotations. Experimental results on a benchmark synthetic dataset and the Pascal3D+ real-world dataset show that our model outperforms the current state-of-the-art methods significantly.

Yijiang Chen, Xiangdong Zhou, Zhen Xing et al.

Self-supervised representation learning of Multivariate Time Series (MTS) is a challenging task and attracts increasing research interests in recent years. Many previous works focus on the pretext task of self-supervised learning and usually neglect the complex problem of MTS encoding, leading to unpromising results. In this paper, we tackle this challenge from two aspects: encoder and pretext task, and propose a unified channel-aware self-supervised learning framework CaSS. Specifically, we first design a new Transformer-based encoder Channel-aware Transformer (CaT) to capture the complex relationships between different time channels of MTS. Second, we combine two novel pretext tasks Next Trend Prediction (NTP) and Contextual Similarity (CS) for the self-supervised representation learning with our proposed encoder. Extensive experiments are conducted on several commonly used benchmark datasets. The experimental results show that our framework achieves new state-of-the-art comparing with previous self-supervised MTS representation learning methods (up to +7.70\% improvement on LSST dataset) and can be well applied to the downstream MTS classification.

Zhidan Liu, Zhen Xing, Xiangdong Zhou et al.

Image-based object pose estimation sounds amazing because in real applications the shape of object is oftentimes not available or not easy to take like photos. Although it is an advantage to some extent, un-explored shape information in 3D vision learning problem looks like "flaws in jade". In this paper, we deal with the problem in a reasonable new setting, namely 3D shape is exploited in the training process, and the testing is still purely image-based. We enhance the performance of image-based methods for category-agnostic object pose estimation by exploiting 3D knowledge learned by a multi-modal method. Specifically, we propose a novel contrastive knowledge distillation framework that effectively transfers 3D-augmented image representation from a multi-modal model to an image-based model. We integrate contrastive learning into the two-stage training procedure of knowledge distillation, which formulates an advanced solution to combine these two approaches for cross-modal tasks. We experimentally report state-of-the-art results compared with existing category-agnostic image-based methods by a large margin (up to +5% improvement on ObjectNet3D dataset), demonstrating the effectiveness of our method.

Ruian He, Zhen Xing, Weimin Tan et al.

Self-supervised representation learning has gained increasing attention for strong generalization ability without relying on paired datasets. However, it has not been explored sufficiently for facial representation. Self-supervised facial representation learning remains unsolved due to the coupling of facial identities, expressions, and external factors like pose and light. Prior methods primarily focus on contrastive learning and pixel-level consistency, leading to limited interpretability and suboptimal performance. In this paper, we propose LatentFace, a novel generative framework for self-supervised facial representations. We suggest that the disentangling problem can be also formulated as generative objectives in space and time, and propose the solution using a 3D-aware latent diffusion model. First, we introduce a 3D-aware autoencoder to encode face images into 3D latent embeddings. Second, we propose a novel representation diffusion model to disentangle 3D latent into facial identity and expression. Consequently, our method achieves state-of-the-art performance in facial expression recognition (FER) and face verification among self-supervised facial representation learning models. Our model achieves a 3.75\% advantage in FER accuracy on RAF-DB and 3.35\% on AffectNet compared to SOTA methods.

Zhaohe Liao, Kaixun Jiang, Zhihang Liu et al.

Although image generation has boosted various applications via its rapid evolution, whether the state-of-the-art models are able to produce ready-to-use academic illustrations for papers is still largely unexplored. Directly comparing or evaluating the illustration with VLM is native but requires oracle multi-modal understanding ability, which is unreliable for long and complex texts and illustrations. To address this, we propose AIBench, the first benchmark using VQA for evaluating logic correctness of the academic illustrations and VLMs for assessing aesthetics. In detail, we designed four levels of questions proposed from a logic diagram summarized from the method part of the paper, which query whether the generated illustration aligns with the paper on different scales. Our VQA-based approach raises more accurate and detailed evaluations on visual-logical consistency while relying less on the ability of the judger VLM. With our high-quality AIBench, we conduct extensive experiments and conclude that the performance gap between models on this task is significantly larger than general ones, reflecting their various complex reasoning and high-density generation ability. Further, the logic and aesthetics are hard to optimize simultaneously as in handcrafted illustrations. Additional experiments further state that test-time scaling on both abilities significantly boosts the performance on this task.

Pengbo Guo, Junke Wang, Zhen Xing et al.

This paper presents DeRA, a novel 1D video tokenizer that decouples the spatial-temporal representation learning in video tokenization to achieve better training efficiency and performance. Specifically, DeRA maintains a compact 1D latent space while factorizing video encoding into appearance and motion streams, which are aligned with pretrained vision foundation models to capture the spatial semantics and temporal dynamics in videos separately. To address the gradient conflicts introduced by the heterogeneous supervision, we further propose the Symmetric Alignment-Conflict Projection (SACP) module that proactively reformulates gradients by suppressing the components along conflicting directions. Extensive experiments demonstrate that DeRA outperforms LARP, the previous state-of-the-art video tokenizer by 25% on UCF-101 in terms of rFVD. Moreover, using DeRA for autoregressive video generation, we also achieve new state-of-the-art results on both UCF-101 class-conditional generation and K600 frame prediction.

Quanhao Li, Junqiu Yu, Kaixun Jiang et al.

Reinforcement learning has emerged as a powerful tool for improving diffusion-based text-to-image models, but existing methods are largely limited to single-task optimization. Extending RL to multiple tasks is challenging: joint optimization suffers from cross-task interference and imbalance, while cascade RL is cumbersome and prone to catastrophic forgetting. We propose DiffusionOPD, a new multi-task training paradigm for diffusion models based on Online Policy Distillation (OPD). DiffusionOPD first trains task-specific teachers independently, then distills their capabilities into a unified student along the student own rollout trajectories. This decouples single-task exploration from multi-task integration and avoids the optimization burden of solving all tasks jointly from scratch. Theoretically, we lift the OPD framework from discrete tokens to continuous-state Markov processes, deriving a closed-form per-step KL objective that unifies both stochastic SDE and deterministic ODE refinement via mean-matching. We formally and empirically demonstrate that this analytic gradient provides lower variance and better generality compared to conventional PPO-style policy gradients. Extensive experiments show that DiffusionOPD consistently surpasses both multi-reward RL and cascade RL baselines in training efficiency and final performance, while achieving state-of-the-art results on all evaluated benchmarks.

Xinda Chen, Zhen Xing, Hanyu Zhang et al.

Tabular data is a fundamental form of data structure. The evolution of table analysis tools reflects humanity's continuous progress in data acquisition, management, and processing. The dynamic changes in table columns arise from technological advancements, changing needs, data integration, etc. However, the standard process of training AI models on tables with fixed columns and then performing inference is not suitable for handling dynamically changed tables. Therefore, new methods are needed for efficiently handling such tables in an unsupervised manner. In this paper, we introduce a new task, Tabular Incremental Inference (TabII), which aims to enable trained models to incorporate new columns during the inference stage, enhancing the practicality of AI models in scenarios where tables are dynamically changed. Furthermore, we demonstrate that this new task can be framed as an optimization problem based on the information bottleneck theory, which emphasizes that the key to an ideal tabular incremental inference approach lies in minimizing mutual information between tabular data and representation while maximizing between representation and task labels. Under this guidance, we design a TabII method with Large Language Model placeholders and Pretrained TabAdapter to provide external knowledge and Incremental Sample Condensation blocks to condense the task-relevant information given by incremental column attributes. Experimental results across eight public datasets show that TabII effectively utilizes incremental attributes, achieving state-of-the-art performance.

Quanhao Li, Zhen Xing, Rui Wang et al.

Recent advances in trajectory-controllable video generation have achieved remarkable progress. Previous methods mainly use adapter-based architectures for precise motion control along predefined trajectories. However, all these methods rely on a multi-step denoising process, leading to substantial time redundancy and computational overhead. While existing video distillation methods successfully distill multi-step generators into few-step, directly applying these approaches to trajectory-controllable video generation results in noticeable degradation in both video quality and trajectory accuracy. To bridge this gap, we introduce FlashMotion, a novel training framework designed for few-step trajectory-controllable video generation. We first train a trajectory adapter on a multi-step video generator for precise trajectory control. Then, we distill the generator into a few-step version to accelerate video generation. Finally, we finetune the adapter using a hybrid strategy that combines diffusion and adversarial objectives, aligning it with the few-step generator to produce high-quality, trajectory-accurate videos. For evaluation, we introduce FlashBench, a benchmark for long-sequence trajectory-controllable video generation that measures both video quality and trajectory accuracy across varying numbers of foreground objects. Experiments on two adapter architectures show that FlashMotion surpasses existing video distillation methods and previous multi-step models in both visual quality and trajectory consistency.

Shuyuan Tu, Yueming Pan, Yinming Huang et al.

Current diffusion-based acceleration methods for long-portrait animation struggle to ensure identity (ID) consistency. This paper presents FlashPortrait, an end-to-end video diffusion transformer capable of synthesizing ID-preserving, infinite-length videos while achieving up to 6x acceleration in inference speed. In particular, FlashPortrait begins by computing the identity-agnostic facial expression features with an off-the-shelf extractor. It then introduces a Normalized Facial Expression Block to align facial features with diffusion latents by normalizing them with their respective means and variances, thereby improving identity stability in facial modeling. During inference, FlashPortrait adopts a dynamic sliding-window scheme with weighted blending in overlapping areas, ensuring smooth transitions and ID consistency in long animations. In each context window, based on the latent variation rate at particular timesteps and the derivative magnitude ratio among diffusion layers, FlashPortrait utilizes higher-order latent derivatives at the current timestep to directly predict latents at future timesteps, thereby skipping several denoising steps and achieving 6x speed acceleration. Experiments on benchmarks show the effectiveness of FlashPortrait both qualitatively and quantitatively.

Ruian He, Zhen Xing, Weimin Tan et al.

Affective Analysis is not a single task, and the valence-arousal value, expression class, and action unit can be predicted at the same time. Previous researches did not pay enough attention to the entanglement and hierarchical relation of these three facial attributes. We propose a novel model named feature pyramid networks for multi-task affect analysis. The hierarchical features are extracted to predict three labels and we apply a teacher-student training strategy to learn from pretrained single-task models. Extensive experiment results demonstrate the proposed model outperforms other models. This is a submission to The 2nd Workshop and Competition on Affective Behavior Analysis in the wild (ABAW). The code and model are available for research purposes at https://github.com/ryanhe312/ABAW2-FPNMAA.

Shuyuan Tu, Zhen Xing, Xintong Han et al.

Current diffusion models for human image animation struggle to ensure identity (ID) consistency. This paper presents StableAnimator, the first end-to-end ID-preserving video diffusion framework, which synthesizes high-quality videos without any post-processing, conditioned on a reference image and a sequence of poses. Building upon a video diffusion model, StableAnimator contains carefully designed modules for both training and inference striving for identity consistency. In particular, StableAnimator begins by computing image and face embeddings with off-the-shelf extractors, respectively and face embeddings are further refined by interacting with image embeddings using a global content-aware Face Encoder. Then, StableAnimator introduces a novel distribution-aware ID Adapter that prevents interference caused by temporal layers while preserving ID via alignment. During inference, we propose a novel Hamilton-Jacobi-Bellman (HJB) equation-based optimization to further enhance the face quality. We demonstrate that solving the HJB equation can be integrated into the diffusion denoising process, and the resulting solution constrains the denoising path and thus benefits ID preservation. Experiments on multiple benchmarks show the effectiveness of StableAnimator both qualitatively and quantitatively.

Ziwei Zhou, Zeyuan Lai, Rui Wang et al.

Text-to-Audio-Video (T2AV) generation is rapidly becoming a core interface for media creation, yet its evaluation remains fragmented. Existing benchmarks largely assess audio and video in isolation or rely on coarse embedding similarity, failing to capture the fine-grained joint correctness required by realistic prompts. We introduce AVGen-Bench, a task-driven benchmark for T2AV generation featuring high-quality prompts across 11 real-world categories. To support comprehensive assessment, we propose a multi-granular evaluation framework that combines lightweight specialist models with Multimodal Large Language Models (MLLMs), enabling evaluation from perceptual quality to fine-grained semantic controllability. Our evaluation reveals a pronounced gap between strong audio-visual aesthetics and weak semantic reliability, including persistent failures in text rendering, speech coherence, physical reasoning, and a universal breakdown in musical pitch control. Code and benchmark resources are available at http://aka.ms/avgenbench.

Quanhao Li, Zhen Xing, Rui Wang et al.

Recent advances in video generation have led to remarkable improvements in visual quality and temporal coherence. Upon this, trajectory-controllable video generation has emerged to enable precise object motion control through explicitly defined spatial paths. However, existing methods struggle with complex object movements and multi-object motion control, resulting in imprecise trajectory adherence, poor object consistency, and compromised visual quality. Furthermore, these methods only support trajectory control in a single format, limiting their applicability in diverse scenarios. Additionally, there is no publicly available dataset or benchmark specifically tailored for trajectory-controllable video generation, hindering robust training and systematic evaluation. To address these challenges, we introduce MagicMotion, a novel image-to-video generation framework that enables trajectory control through three levels of conditions from dense to sparse: masks, bounding boxes, and sparse boxes. Given an input image and trajectories, MagicMotion seamlessly animates objects along defined trajectories while maintaining object consistency and visual quality. Furthermore, we present MagicData, a large-scale trajectory-controlled video dataset, along with an automated pipeline for annotation and filtering. We also introduce MagicBench, a comprehensive benchmark that assesses both video quality and trajectory control accuracy across different numbers of objects. Extensive experiments demonstrate that MagicMotion outperforms previous methods across various metrics. Our project page are publicly available at https://quanhaol.github.io/magicmotion-site.

Shuyuan Tu, Zhen Xing, Xintong Han et al.

Current diffusion models for human image animation often struggle to maintain identity (ID) consistency, especially when the reference image and driving video differ significantly in body size or position. We introduce StableAnimator++, the first ID-preserving video diffusion framework with learnable pose alignment, capable of generating high-quality videos conditioned on a reference image and a pose sequence without any post-processing. Building upon a video diffusion model, StableAnimator++ contains carefully designed modules for both training and inference, striving for identity consistency. In particular, StableAnimator++ first uses learnable layers to predict the similarity transformation matrices between the reference image and the driven poses via injecting guidance from Singular Value Decomposition (SVD). These matrices align the driven poses with the reference image, mitigating misalignment to a great extent. StableAnimator++ then computes image and face embeddings using off-the-shelf encoders, refining the face embeddings via a global content-aware Face Encoder. To further maintain ID, we introduce a distribution-aware ID Adapter that counteracts interference caused by temporal layers while preserving ID via distribution alignment. During the inference stage, we propose a novel Hamilton-Jacobi-Bellman (HJB) based face optimization integrated into the denoising process, guiding the diffusion trajectory for enhanced facial fidelity. Experiments on benchmarks show the effectiveness of StableAnimator++ both qualitatively and quantitatively.

Qijun Feng, Zhen Xing, Zuxuan Wu et al.

We introduce GeoGS3D, a novel two-stage framework for reconstructing detailed 3D objects from single-view images. Inspired by the success of pre-trained 2D diffusion models, our method incorporates an orthogonal plane decomposition mechanism to extract 3D geometric features from the 2D input, facilitating the generation of multi-view consistent images. During the following Gaussian Splatting, these images are fused with epipolar attention, fully utilizing the geometric correlations across views. Moreover, we propose a novel metric, Gaussian Divergence Significance (GDS), to prune unnecessary operations during optimization, significantly accelerating the reconstruction process. Extensive experiments demonstrate that GeoGS3D generates images with high consistency across views and reconstructs high-quality 3D objects, both qualitatively and quantitatively.

Shuyuan Tu, Yueming Pan, Yinming Huang et al.

Current diffusion models for audio-driven avatar video generation struggle to synthesize long videos with natural audio synchronization and identity consistency. This paper presents StableAvatar, the first end-to-end video diffusion transformer that synthesizes infinite-length high-quality videos without post-processing. Conditioned on a reference image and audio, StableAvatar integrates tailored training and inference modules to enable infinite-length video generation. We observe that the main reason preventing existing models from generating long videos lies in their audio modeling. They typically rely on third-party off-the-shelf extractors to obtain audio embeddings, which are then directly injected into the diffusion model via cross-attention. Since current diffusion backbones lack any audio-related priors, this approach causes severe latent distribution error accumulation across video clips, leading the latent distribution of subsequent segments to drift away from the optimal distribution gradually. To address this, StableAvatar introduces a novel Time-step-aware Audio Adapter that prevents error accumulation via time-step-aware modulation. During inference, we propose a novel Audio Native Guidance Mechanism to further enhance the audio synchronization by leveraging the diffusion's own evolving joint audio-latent prediction as a dynamic guidance signal. To enhance the smoothness of the infinite-length videos, we introduce a Dynamic Weighted Sliding-window Strategy that fuses latent over time. Experiments on benchmarks show the effectiveness of StableAvatar both qualitatively and quantitatively.

Hanyu Zhang, Zhen Xing, Ruian He et al.

Coreset selection aims to identify a small yet highly informative subset of data, thereby enabling more efficient model training while reducing storage overhead. Recently, this capability has been leveraged to tackle the challenges of fine-tuning large foundation models, offering a direct pathway to their efficient and practical deployment. However, most existing methods are class-agnostic, causing them to overlook significant difficulty variations among classes. This leads them to disproportionately prune samples from either overly easy or hard classes, resulting in a suboptimal allocation of the data budget that ultimately degrades the final coreset performance. To address this limitation, we propose Non-Uniform Class-Wise Coreset Selection (NUCS), a novel framework that both integrates class-level and sample-level difficulty. We propose a robust metric for global class difficulty, quantified as the winsorized average of per-sample difficulty scores. Guided by this metric, our method performs a theoretically-grounded, non-uniform allocation of data selection budgets inter-class, while adaptively selecting samples intra-class with optimal difficulty ranges. Extensive experiments on a wide range of visual classification tasks demonstrate that NUCS consistently outperforms state-of-the-art methods across 10 diverse datasets and pre-trained models, achieving both superior accuracy and computational efficiency, highlighting the promise of non-uniform class-wise selection strategy for advancing the efficient fine-tuning of large foundation models.

Miaosen Zhang, Yixuan Wei, Zhen Xing et al.

Modern vision models are trained on very large noisy datasets. While these models acquire strong capabilities, they may not follow the user's intent to output the desired results in certain aspects, e.g., visual aesthetic, preferred style, and responsibility. In this paper, we target the realm of visual aesthetics and aim to align vision models with human aesthetic standards in a retrieval system. Advanced retrieval systems usually adopt a cascade of aesthetic models as re-rankers or filters, which are limited to low-level features like saturation and perform poorly when stylistic, cultural or knowledge contexts are involved. We find that utilizing the reasoning ability of large language models (LLMs) to rephrase the search query and extend the aesthetic expectations can make up for this shortcoming. Based on the above findings, we propose a preference-based reinforcement learning method that fine-tunes the vision models to distill the knowledge from both LLMs reasoning and the aesthetic models to better align the vision models with human aesthetics. Meanwhile, with rare benchmarks designed for evaluating retrieval systems, we leverage large multi-modality model (LMM) to evaluate the aesthetic performance with their strong abilities. As aesthetic assessment is one of the most subjective tasks, to validate the robustness of LMM, we further propose a novel dataset named HPIR to benchmark the alignment with human aesthetics. Experiments demonstrate that our method significantly enhances the aesthetic behaviors of the vision models, under several metrics. We believe the proposed algorithm can be a general practice for aligning vision models with human values.

Zhen Xing, Qi Dai, Zejia Weng et al.

Text-guided video prediction (TVP) involves predicting the motion of future frames from the initial frame according to an instruction, which has wide applications in virtual reality, robotics, and content creation. Previous TVP methods make significant breakthroughs by adapting Stable Diffusion for this task. However, they struggle with frame consistency and temporal stability primarily due to the limited scale of video datasets. We observe that pretrained Image2Video diffusion models possess good priors for video dynamics but they lack textual control. Hence, transferring Image2Video models to leverage their video dynamic priors while injecting instruction control to generate controllable videos is both a meaningful and challenging task. To achieve this, we introduce the Multi-Modal Large Language Model (MLLM) to predict future video states based on initial frames and text instructions. More specifically, we design a dual query transformer (DQFormer) architecture, which integrates the instructions and frames into the conditional embeddings for future frame prediction. Additionally, we develop Long-Short Term Temporal Adapters and Spatial Adapters that can quickly transfer general video diffusion models to specific scenarios with minimal training costs. Experimental results show that our method significantly outperforms state-of-the-art techniques on four datasets: Something Something V2, Epic Kitchen-100, Bridge Data, and UCF-101. Notably, AID achieves 91.2% and 55.5% FVD improvements on Bridge and SSv2 respectively, demonstrating its effectiveness in various domains. More examples can be found at our website https://chenhsing.github.io/AID.

Bei Li, Yi Jing, Xu Tan et al.

Learning multiscale Transformer models has been evidenced as a viable approach to augmenting machine translation systems. Prior research has primarily focused on treating subwords as basic units in developing such systems. However, the incorporation of fine-grained character-level features into multiscale Transformer has not yet been explored. In this work, we present a \textbf{S}low-\textbf{F}ast two-stream learning model, referred to as Tran\textbf{SF}ormer, which utilizes a ``slow'' branch to deal with subword sequences and a ``fast'' branch to deal with longer character sequences. This model is efficient since the fast branch is very lightweight by reducing the model width, and yet provides useful fine-grained features for the slow branch. Our TranSFormer shows consistent BLEU improvements (larger than 1 BLEU point) on several machine translation benchmarks.