论文检索

Mingxing Rao, Bowen Qu, Daniel Moyer

The recovery of training data from generative models ("model inversion") has been extensively studied for diffusion models in the data domain as a memorization/overfitting phenomenon. Latent diffusion models (LDMs), which operate on the latent codes from encoder/decoder pairs, have been robust to prior inversion methods. In this work we describe two key findings: (1) the diffusion model exhibits non-uniform memorization across latent codes, tending to overfit samples located in high-distortion regions of the decoder pullback metric; (2) even within a single latent code, memorization contributions are unequal across representation dimensions. Our proposed method to ranks latent dimensions by their contribution to the decoder pullback metric, which in turn identifies dimensions that contribute to memorization. For score-based membership inference, a sub-task of model inversion, we find that removing less-memorizing dimensions improves performance on all tested methods and datasets, with average AUROC gains of 1-4% and substantial increases in TPR@1%FPR (1-32%) across diverse datasets including CIFAR-10, CelebA, ImageNet-1K, Pokemon, MS-COCO, and Flickr. Our results highlight the overlooked influence of the auto-encoder geometry on LDM memorization and provide a new perspective for analyzing privacy risks in diffusion-based generative models.

Caleb Zheng, Eli Shlizerman

Diffusion models achieve remarkable performance across diverse generative tasks in computer vision, but their high computational cost remains a major barrier to deployment. Model pruning offers a promising way to reduce inference cost and enable lightweight models. However, pruning leads to quality drop due to reduced capacity. A key limitation of existing pruning approaches is that pruned models are finetuned using the same objective as the dense model (denoising score matching). Since the dense model is accessible during finetuning, it warrants a more effective approach for knowledge transfer from the dense to the pruned model. Motivated by this, we propose 2ndMatch(2ndM), a general-purpose finetuning framework that introduces a 2nd-order Jacobian (J^ \top J) Matching loss inspired by Finite-Time Lyapunov Exponents. 2ndM teaches the pruned model to mimic the sensitivity of the dense teacher, i.e., how to respond to small perturbations over time, through scalable random projections. The framework is architecture-agnostic and applies to both U-Net- and Transformer-based diffusion models. Experiments on CIFAR-10, CelebA, LSUN, ImageNet, and MSCOCO demonstrate that 2ndM reduces the performance gap between pruned and dense models, substantially improving output quality.

Ziyue Lin, Jiahe Hou, Hongyu Xia, Xinrui Xie, Feifei Wang, Yuyin Zhou, Wei Wang, Jiawei Liu, Liangqiong Qu

We propose Decoupled Residual Denoising Diffusion models (DRDD) for unified and data-efficient image-to-image (I2I) translation. While diffusion models have advanced I2I translation in terms of quality and diversity, we uncover a previously under-explored property in diffusion models. Crucially, beyond its conventional role of manifold lifting (i.e., moving data off low-dimensional manifolds), injecting Gaussian noise facilitates domain harmonization by implicitly aligning feature distributions across domains, a property particularly advantageous for unified I2I translation. However, existing diffusion models prematurely erode this harmonization effect, as noise and residuals are simultaneously removed in a single coupled diffusion process. To address this, DRDD decouples the diffusion process into two sequential and independent diffusion stages: (1) a stochastic noise diffusion for domain harmonization and manifold lifting, and (2) a deterministic residual diffusion that learns the core semantic mapping entirely within the fixed-noise domain. This decoupling preserves harmonization and manifold lifting effects throughout the transformation, substantially simplifying the learning of unified mappings across diverse tasks and domains. Notably, the noise diffusion stage is trained exclusively on abundant, unpaired target-domain images, greatly improving data efficiency. Comprehensive theoretical and empirical analysis demonstrates that DRDD is broadly compatible with mainstream diffusion models and consistently delivers robust, unified I2I translation, even under limited paired data. Our code is available at https://github.com/HKU-HealthAI/DRDD.

Guohao Zhao, Yuxin Peng

Virtual try-on (VTON) aims to render a target garment onto a person while preserving pose, identity, and fine-grained appearance. Most existing methods rely on supervised paired data, limiting cross-domain generalization, while recent training-free approaches, though more robust, require multiple diffusion calls and complex compositing, making deployment impractical. We propose PG-VTON, a single-pass, training-free framework based on Patch-Guided Reference Alignment. Our key insight is that modern inpainting diffusion models already possess strong in-context completion: given a masked person and a small garment patch, they can synthesize plausible, pose-consistent clothing without task-specific training. PG-VTON exploits this capability with two lightweight components: Patch-Anchored Identity Priming (PIP) injects a localized garment patch only in early denoising steps to anchor garment identity, and Reference-Aware Attention (RAA) strengthens attention from masked-region tokens to garment tokens to enhance detail transfer, all without modifying model weights. With a single diffusion pass, PG-VTON achieves state-of-the-art performance among training-free methods on DressCode and VITON-HD and generalizes effectively to subject insertion. Code is available at \href https://github.com/PKU-ICST-MIPL/PG-VTON_CVPR2026 https://github.com/PKU-ICST-MIPL/PG-VTON_CVPR2026 .

Chenru Wang, Yunyi Chen, Zijun Yang, Joey Tianyi Zhou, Chi Zhang

Dataset Distillation aims to synthesize compact datasets that can approximate the training efficacy of large-scale real datasets, offering an efficient solution to the increasing computational demands of modern deep learning. Recently, diffusion-based dataset distillation methods have shown great promise by leveraging the strong generative capacity of diffusion models to produce diverse and structurally consistent samples. However, a fundamental goal misalignment persists: diffusion models are optimized for generative likelihood rather than discriminative utility, resulting in over-concentration in high-density regions and inadequate coverage of boundary samples crucial for classification. To address this issue, we propose two complementary strategies. Inversion-Matching (IM) introduces an inversion-guided fine-tuning process that aligns denoising trajectories with their inversion counterparts, broadening distributional coverage and enhancing diversity. Selective Subgroup Sampling( S^3 ) is a training-free sampling mechanism that improves inter-class separability by selecting synthetic subsets that are both representative and distinctive. Extensive experiments demonstrate that our approach significantly enhances the discriminative quality and generalization of distilled datasets, achieving state-of-the-art performance among diffusion-based methods.

Ziqi Cai, Taoyu Yang, Zheng Chang, Si Li, Han Jiang, Shuchen Weng, Boxin Shi

Diffusion models have achieved remarkable progress in video generation, but their controllability remains a major limitation. Key scene factors such as layout, lighting, and camera trajectory are often entangled or only weakly modeled, restricting their applicability in domains like filmmaking and virtual production where explicit scene control is essential. We present LiVER, a diffusion-based framework for scene-controllable video generation. To achieve this, we introduce a novel framework that conditions video synthesis on explicit 3D scene properties, supported by a new large-scale dataset with dense annotations of object layout, lighting, and camera parameters. Our method disentangles these properties by rendering control signals from a unified 3D representation. We propose a lightweight conditioning module and a progressive training strategy to integrate these signals into a foundational video diffusion model, ensuring stable convergence and high fidelity. Our framework enables a wide range of applications, including image-to-video and video-to-video synthesis where the underlying 3D scene is fully editable. To further enhance usability, we develop a scene agent that automatically translates high-level user instructions into the required 3D control signals. Experiments show that LiVER achieves state-of-the-art photorealism and temporal consistency while enabling precise, disentangled control over scene factors, setting a new standard for controllable video generation.

Xiaojun Deng, Tianchi Liao, Zhiyuan Liu, Chuan Chen, Zibin Zheng

One-shot Federated Learning (OSFL) has emerged as a promising paradigm to mitigate the high communication overhead of traditional federated learning. However, its effectiveness is often hindered by data heterogeneity across clients. While recent methods leverage pre-trained diffusion models to generate data for OSFL, they often struggle with some practical limitations, including a lack of semantic fidelity in capturing the fine-grained characteristics of local data, and insufficient diversity in the generated data, which collectively degrade the performance of the global model. To address these challenges, we propose Espresso, a novel framework that enhances both the fidelity and diversity of synthetic data in OSFL. Espresso consists of two main components: (1) Fine-Grained Condition Learning, which learns fine-grained conditional embeddings to improve semantic fidelity and diversity by modeling intra-category patterns, and (2) Semantics-Preserving Sampling, which diversifies the generated data by modeling the distribution of latent noises and applying a self-reflection sampling strategy. Extensive experiments on benchmark datasets demonstrate that Espresso can improve the semantic fidelity and diversity of the synthetic data, leading to an enhancement in the performance of the global model compared to state-of-the-art OSFL methods under data heterogeneity.

Yunhong Lu, Yanhong Zeng, Haobo Li, Hao Ouyang, Qiuyu Wang, Ka Leong Cheng, Jiapeng Zhu, Hengyuan Cao, Zhipeng Zhang, Xing Zhu 等

Efficient streaming video generation is critical for simulating interactive and dynamic worlds. Existing methods distill few-step video diffusion models with sliding window attention, using initial frames as sink tokens to maintain attention performance and reduce error accumulation. However, video frames become overly dependent on these static tokens, resulting in copied initial frames and diminished motion dynamics. To address this, we introduce Reward Forcing, a novel framework with two key designs. First, we propose EMA-Sink, which maintains fixed-size tokens initialized from initial frames and continuously updated by fusing evicted tokens via exponential moving average as they exit the sliding window. Without additional computation cost, EMA-Sink tokens capture both long-term context and recent dynamics, preventing initial frame copying while maintaining long-horizon consistency. Second, to better distill motion dynamics from teacher models, we propose a novel Rewarded Distribution Matching Distillation (Re-DMD). Vanilla distribution matching treats every training sample equally, limiting the model's ability to prioritize dynamic content. Instead, Re-DMD biases the model's output distribution toward high-reward regions by prioritizing samples with greater dynamics rated by a vision-language model. Re-DMD significantly enhances motion quality while preserving data fidelity. We include both quantitative and qualitative experiments to show that Reward Forcing achieves state-of-the-art performance on standard benchmarks while enabling high-quality streaming video generation at 23.1 FPS on a single H100 GPU.

Ziming Liu, Yifan Yang, Chengruidong Zhang, Yiqi Zhang, Lili Qiu, Yang You, Yuqing Yang

Diffusion models (DMs) have become the state-of-the-art for generative tasks across domains, but their reliance on sequential forward passes limits real-time performance. Prior acceleration methods mainly reduce sampling steps or reuse intermediate results. Leveraging the flexibility of Diffusion Transformers (DiTs) to handle variable token counts, we propose RAS, a training-free sampling strategy that dynamically assigns different update ratios to image regions based on model focus. Our key observation is that at each step, DiTs concentrate on semantically meaningful areas, and these regions exhibit strong continuity across consecutive steps. Exploiting this, RAS updates only focused regions while reusing cached noise for others, with focus determined from the previous step's output. Evaluated on Stable Diffusion 3 and Lumina-Next-T2I, RAS achieves up to 2.36x and 2.51x speedups, respectively, with minimal quality loss. This demonstrates a practical step toward more efficient diffusion transformers for real-time generation.

Tuan Dung Nguyen, Minh Khoi Ho, Qi Chen, Yutong Xie, Cam-Tu Nguyen, Minh Khoi Nguyen, Dang Huy Pham Nguyen, Anton van den Hengel, Johan Verjans, Phi Le Nguyen 等

Large vision-language models (LVLMs) achieve strong performance on visual reasoning tasks but remain highly susceptible to hallucination. Existing detection methods predominantly rely on coarse, whole-image measures of how an object token relates to the input image. This global strategy is limited: hallucinated tokens may exhibit weak but widely scattered correlations across many local regions, which aggregate into deceptively high overall relevance, thus evading the current global hallucination detectors. We begin with a simple yet critical observation: a faithful object token must be strongly grounded in a specific image region. Building on this insight, we introduce a patch-level hallucination detection framework that examines fine-grained token-level interactions across model layers. Our analysis uncovers two characteristic signatures of hallucinated tokens: (i) they yield diffuse, non-localized attention patterns, in contrast to the compact, well-focused attention and (ii) they fail to exhibit meaningful semantic alignment with any visual region. Guided by these findings, we develop a lightweight and interpretable detection method that leverages patch-level statistical features, combined with hidden-layer representations. Our approach achieves up to 90% accuracy in token-level hallucination detection, demonstrating the superiority of fine-grained structural analysis for detecting hallucinations.

Min Shi, Xiaohui Zeng, Jiannan Huang, Yin Cui, Francesco Ferroni, Jialuo Li, Zhaoshuo Li, Yogesh Balaji, Haoxiang Wang, Tsung-Yi Lin 等

Interleaved multimodal generation enables capabilities beyond unimodal generation models, such as step-by-step instructional guides, visual planning, and generating visual drafts for reasoning. However, the quality of existing interleaved generation models under general instructions remains limited by insufficient training data and base model capacity. We present DuoGen, an interleaved generation framework that systematically addresses data curation, architecture design, and evaluation. On the data side, we build a large-scale, high-quality instruction-tuning dataset by combining multimodal conversations rewritten from curated raw websites, and diverse synthetic examples covering everyday scenarios. Architecturally, DuoGen leverages the strong visual understanding of a pretrained multimodal LLM and the visual generation capabilities of a diffusion transformer (DiT) pretrained on video generation, avoiding costly unimodal pretraining and enabling flexible base model selection. A two-stage decoupled strategy first instruction-tunes the MLLM, then aligns DiT with it using curated interleaved image-text sequences. Across public and newly proposed benchmarks, DuoGen outperforms prior open-source models in text quality, image fidelity, and image-context alignment, and also achieves state-of-the-art performance on text-to-image and image editing among unified generation models. Data and code are released at https://research.nvidia.com/labs/dir/duogen/

Ziyang Wang, Yue Zhang, Mingdao Wang, Yasen Zhang, Teer Song, Yu Tian, Xueming Li

Visual Autoregressive Modeling (VAR) has recently emerged as a powerful paradigm for image generation that surpasses diffusion models in efficiency and quality. However, accelerating attention computation in VAR is still challenging because attention patterns across scales exhibit strong and complex semantic biases that early coarse-scale tokens dominate global structure, while fine-scale tokens mainly refine local details. Existing acceleration methods rely on heuristic token pruning or fixed attention masks, lacking a principled way to balance acceleration and semantic fidelity. In this work, we propose a two-stage acceleration framework for VAR. First, we introduce a semantic-cost-aware masking strategy (SCA-Mask) that quantifies the importance of each attention tile and formulates mask shape design as a cost-constrained optimization problem. This enables adaptive pruning under a given compute budget while preserving essential semantic context. Second, we present Post-Acceleration Adaptation (PAA), a decoder-side fine-tuning scheme that employs internal knowledge distillation to restore image quality from pruned latents. PAA does not require external data and uses a lightweight LoRA-based adaptation, providing a highly efficient alternative to retraining the autoregressive transformer. Comprehensive experiments across multiple VAR tasks demonstrate that our method achieves decent speedup with negligible loss of visual fidelity, yielding a principled and effective pathway toward fast and high-quality visual autoregressive generation.

Zeyu An, Wanyu Lin, Feng Tan, Shujun Wang

Recent advances in diffusion-based language models (DLMs) have shown remarkable potential for de novo protein design. However, enabling controllable protein generation requires integrating diverse biological conditions, such as structure, functions, and chemical interactions, each represented in distinct modalities. Existing approaches often either support a single condition or treat multiple conditions through separate modality-specific encoders. This isolation limits cross-modal interaction, reduces generation quality, and complicates the incorporation of new conditions without retraining or redesigning the backbone. To address these limitations, we introduce MMCP-GEN, a DLM for Multi-Modal, Multi-Condition Protein sequence GENeration. MMCP-GEN establishes a new paradigm for controllable protein generation under complex multimodal constraints. Its core is a modality-composable and extensible conditioning mechanism that fuses heterogeneous biological conditions via learnable queries and modality-indicator heads, enabling disentangled, extensible, and cross-modal condition integration without retraining the backbone. A joint generation-and-scoring objective further aligns sequence recovery with structural fidelity. Empirically, MMCP-GEN achieves state-of-the-art performance across structure-, function-, and ligand-conditioned tasks, improving sequence recovery by up to 5% and outperforming attentive baselines in diverse functional annotation tasks. These results establish MMCP-GEN as a general and high-fidelity framework for controllable protein generation. The source code is publicly available at https://github.com/WanyuGroup.

Yiyang Wang, Xi Chen, Xiaogang Xu, Yu Liu, Hengshuang Zhao

Recent advancements adopt online reinforcement learning (RL) from LLMs to text-to-image rectified flow diffusion models for reward alignment. The use of group-level rewards successfully aligns the model with the targeted reward. However, it faces challenges including low efficiency, dependency on stochastic samplers, and reward hacking. The problem is that rectified flow models are fundamentally different from LLMs: 1) For efficiency, online image sampling takes much more time and dominates the time of training. 2) For stochasticity, rectified flow is deterministic once the initial noise is fixed. Aiming at these problems and inspired by the effects of group-level rewards from LLMs, we design Group-level Direct Reward Optimization (GDRO). GDRO is a new post-training paradigm for group-level reward alignment that combines the characteristics of rectified flow models. Through rigorous theoretical analysis, we point out that GDRO supports full offline training that saves the large time cost for image rollout sampling. Also, it is diffusion-sampler-independent, which eliminates the need for the ODE-to-SDE approximation to obtain stochasticity. We also empirically study the reward hacking trap that may mislead the evaluation, and involve this factor in the evaluation using a corrected score that not only considers the original evaluation reward but also the trend of reward hacking. Extensive experiments demonstrate that GDRO effectively and efficiently improves the reward score of the diffusion model through group-wise offline optimization across the OCR and GenEval tasks, while demonstrating strong stability and robustness in mitigating reward hacking.

Tianyu Chen, Wei Xiang, Kang Han, Yu Lu, Di Wu, Gaowen Liu, Ramana Rao Kompella

Feed-forward 3D reconstruction offers substantial runtime advantages over per-scene optimization, which remains slow at inference and often fragile under sparse views. However, existing feed-forward methods still have potential for further performance gains, especially for out-of-domain data, and struggle to retain second-level inference time once a generative prior is introduced. These limitations stem from the one-shot prediction paradigm in existing feed-forward pipeline: models are strictly bounded by capacity, lack inference-time refinement, and are ill-suited for continuously injecting generative priors. We introduce GIFSplat, a purely feed-forward iterative refinement framework for 3D Gaussian Splatting from sparse unposed views. A small number of forward-only residual updates progressively refine current 3D scene using rendering evidence, achieve favorable balance between efficiency and quality. Furthermore, we distill a frozen diffusion prior into Gaussian-level cues from enhanced novel renderings without gradient backpropagation or ever-increasing view-set expansion, thereby enabling per-scene adaptation with generative prior while preserving feed-forward efficiency. Across DL3DV, RealEstate10K, and DTU, GIFSplat consistently outperforms state-of-the-art feed-forward baselines, improving PSNR by up to +2.1 dB, and it maintains second-scale inference time without requiring camera poses or any test-time gradient optimization. The project page can be found at: https://terencepp.github.io/gifsplat-project-page/.

Quanhao Li, Zhen Xing, Rui Wang, Haidong Cao, Qi Dai, Daoguo Dong, Zuxuan Wu

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.

Fengyi Fang, Sicheng Yang, Wenming Yang

Co-speech gesture generation has significantly advanced human-computer interaction, yet speaker movements remain constrained due to the omission of text-driven non-spontaneous gestures (e.g., bowing while talking). Existing methods face two key challenges: 1) the semantic prior gap due to the lack of descriptive text annotations in gesture datasets, and 2) the difficulty in achieving coordinated multimodal control over gesture generation. To address these challenges, this paper introduces CoordSpeaker, a comprehensive framework that enables coordinated caption-empowered co-speech gesture synthesis. Our approach first bridges the semantic prior gap through a novel gesture captioning framework, leveraging a motion-language model to generate descriptive captions at multiple granularities. Building upon this, we propose a conditional latent diffusion model with unified cross-dataset motion representation and a hierarchically controlled denoiser to achieve highly controlled, coordinated gesture generation. CoordSpeaker pioneers the first exploration of gesture understanding to tackle the semantic gap in gesture generation while offering a novel perspective of bidirectional gesture-text mapping. Extensive experiments demonstrate that our method produces high-quality gestures that are both rhythmically synchronized with speeches and semantically coherent with arbitrary captions, achieving superior performance with higher efficiency compared to existing approaches.

Wenhao Guo, Zhaoran Zhao, Peng Lu, Sheng Li, Qian Qiao, RuiDe Li

Arbitrary-Scale SR (ASISR) remains fundamentally limited by cross-scale distribution shift: once the inference scale leaves the training range, noise, blur, and artifacts accumulate sharply. We revisit this challenge from a cross-scale distribution transition perspective and propose CASR, a simple yet highly efficient cyclic SR framework that reformulates ultra-magnification as a sequence of in-distribution scale transitions. This design ensures stable inference at arbitrary scales while requiring only a single model. CASR tackles two major bottlenecks: distribution drift across iterations and patch-wise diffusion inconsistencies. The proposed SSAM module aligns structural distributions via superpixel aggregation, preventing error accumulation, while SARM module restores high-frequency textures by enforcing correlation-guided consistency and preserving self-similarity structure through correlation alignment. Despite using only a single model, our approach significantly reduces distribution drift, preserves long-range texture consistency, and achieves superior generalization even at extreme magnification.

Christophe Bolduc, Julien Philip, Li Ma, Mingming He, Paul Debevec, Jean-François Lalonde

We present Lighting in Motion (LiMo), a diffusion-based approach to spatiotemporal lighting estimation. LiMo targets both realistic high-frequency detail prediction and accurate illuminance estimation. To account for both, we propose generating a set of mirrored and diffuse spheres at different exposures, based on their 3D positions in the input. Making use of diffusion priors, we fine-tune powerful existing diffusion models on a large-scale customized dataset of indoor and outdoor scenes, paired with spatiotemporal light probes. For accurate spatial conditioning, we demonstrate that depth alone is insufficient and we introduce a new geometric condition to provide the relative position of the scene to the target 3D position. Finally, we combine diffuse and mirror predictions at different exposures into a single HDRI map leveraging differentiable rendering.We thoroughly evaluate our method and design choices to establish LiMo as state-of-the-art for both spatial control and prediction accuracy.

Yannan He, Garvita Tiwari, Xiaohan Zhang, Pankaj Bora, Tolga Birdal, Jan Eric Lenssen, Gerard Pons-Moll

We introduce MoLingo, a text-to-motion (T2M) model that generates realistic, lifelike human motion by denoising in a continuous latent space. Recent works perform latent space diffusion, either on the whole latent at once or auto-regressively over multiple latents. In this paper, we study how to make diffusion on continuous motion latents work best. We focus on two questions: (1) how to build a semantically aligned latent space so diffusion becomes more effective, and (2) how to best inject text conditioning so the motion follows the description closely. We propose a semantic-aligned motion encoder trained with frame-level text labels so that latents with similar text meaning stay close, which makes the latent space more diffusion-friendly. We also compare single-token conditioning with a multi-token cross-attention scheme and find that cross-attention gives better motion realism and text-motion alignment. With semantically aligned latents, auto-regressive generation, and cross-attention text conditioning, our model sets a new state-of-the-art in human motion generation on standard metrics and in a user study. We will release our code and models for further research and downstream usage.