Flow Matching (FM) is an efficient generative modeling framework, but aligning it with human preferences remains underexplored.~Although applying Direct Preference Optimization (DPO) to diffusion models has yielded improvements, directly extending DPO-like methods to FM poses three challenges: 1) Incompatibility with ODE-based models, 2) Heavy computational cost from full model fine-tuning, and 3) Reliance on reference model quality. To address these limitations, we propose Preference Classifier for Flow Matching (PC-Flow), a novel reference-free preference alignment framework. Specifically, we reinterpret FM’s deterministic ODE as an equivalent SDE to enable DPO-style learning. Then, we introduce a lightweight classifier to model relative preferences exclusively. This approach decouples alignment from the generative model, eliminating the need for costly fine-tuning or a reference model. Theoretically, PC-Flow guarantees consistent preference-guided distribution evolution, achieves a DPO-equivalent objective without a reference model, and progressively steers generation toward preferred outputs. Experiments show that PC-Flow achieves DPO-level alignment with significantly lower training costs.
论文检索
输入标题、作者或关键词，从 9,256 篇学术成果中精准定位
Real-time, high-fidelity monocular depth estimation from remote sensing imagery is crucial for numerous applications, yet existing methods face a stark trade-off between accuracy and efficiency. Although using Vision Transformer (ViT) backbones for dense prediction is fast, they often exhibit poor perceptual quality. Conversely, diffusion models offer high fidelity but at a prohibitive computational cost. To overcome these limitations, we propose Depth Detail Diffusion for Remote Sensing Monocular Depth Estimation (D³-RSMDE), an efficient framework designed to achieve an optimal balance between speed and quality. Our framework first leverages a ViT-based module to rapidly generate a high-quality preliminary depth map construction, which serves as a structural prior, effectively replacing the time-consuming initial structure generation stage of diffusion models. Based on this prior, we propose a Progressive Linear Blending Refinement (PLBR) strategy, which uses a lightweight U-Net to refine the details in only a few iterations. The entire refinement step operates efficiently in a compact latent space supported by a Variational Autoencoder (VAE). Extensive experiments demonstrate that D³-RSMDE achieves a notable 11.85% reduction in the Learned Perceptual Image Patch Similarity (LPIPS) perceptual metric over leading models like Marigold, while also achieving over a 40× speedup in inference and maintaining VRAM usage comparable to lightweight ViT models.
Bokeh is used in photography to emphasize the selected subject by smoothly blurring the out-of-focus region with appealing highlights. While recent advances have achieved impressive results in rendering realistic blur, existing frameworks typically rely on disparity maps and bokeh-relevant inputs (e.g., focal distance and blur size), and face significant challenges in video bokeh rendering due to limited temporal consistency. In this paper, we propose BokehCrafter, the first video diffusion framework that generates temporally coherent and visually pleasing bokeh effects from all-in-focus video inputs under user-friendly input conditions. Specifically, we leverage a dual-stream attention mechanism, integrating a reference image branch and a rendering instruction branch. We propose a Bokeh Image Extraction (BIE) module and a CLIP-based text encoder to extract image and text features, respectively, whose outputs are fused via a Text-Image Fusion (TIF) module to enable fine-grained and controllable bokeh rendering. To support the novel capabilities of our model, we construct Video Bokeh Scenes (VBS), a large-scale dataset containing a wide variety of bokeh videos with corresponding rendering instructions, across various scenes and rendering settings. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art methods in both bokeh rendering quality and temporal consistency.
Purple flare, a diffuse chromatic aberration artifact commonly found around highlight areas, severely degrades the tone transition and color of the image. Existing traditional methods are based on hand-crafted features, which lack flexibility and rely entirely on fixed priors, while the scarcity of paired training data critically hampers deep learning. To address this issue, we propose a novel network built upon decoupled HSV Look-Up Tables (LUTs). The method aims to simplify color correction by adjusting the Hue (H), Saturation (S), and Value (V) components independently. This approach resolves the inherent color coupling problems in traditional methods. Our model adopts a two-stage architecture: First, a Chroma-Aware Spectral Tokenizer (CAST) converts the input image from RGB space to HSV space and independently encodes the Hue (H) and Value (V) channels into a set of semantic tokens describing the Purple flare status; second, the HSV-LUT module takes these tokens as input and dynamically generates independent correction curves (1D-LUTs) for the three channels H, S, and V. To effectively train and validate our model, we built the first large-scale purple flare dataset with diverse scenes. We also proposed new metrics and a loss function specifically designed for this task. Extensive experiments demonstrate that our model not only significantly outperforms existing methods in visual effects but also achieves state-of-the-art performance on all quantitative metrics.
Accurately simulating existing 3D objects and a wide variety of materials often demands expert knowledge and time-consuming physical parameter tuning to achieve the desired dynamic behavior. We introduce MotionPhysics, an end‑to‑end differentiable framework that infers plausible physical parameters from a user-proved natural language prompt for a chosen 3D scene of interest, removing the need for guidance from ground‑truth trajectories or annotated videos. Our approach first utilizes a multimodal large language model to estimate material parameter values, which are constrained to be within plausible ranges. We further propose a learnable motion distillation loss, which extracts robust motion priors from pretrained video diffusion models while minimizing appearance and geometry inductive biases to guide the simulation. We evaluate MotionPhysics across more than thirty scenarios, including real-world, human-designed, and AI-generated 3D objects, spanning a wide range of materials such as elastic solids, metals, foams, sand, and both Newtonian and non-Newtonian fluids. We demonstrate that it produces visually realistic dynamic simulations guided by natural language, surpassing the state of the art, with physically plausible parameters that are automatically determined.
FantasyTalking2: Timestep-Layer Adaptive Preference Optimization for Audio-Driven Portrait Animation
PDF ↗Recent advances in audio-driven portrait animation have demonstrated impressive capabilities. However, existing methods struggle to align with fine-grained human preferences across multiple dimensions, such as motion naturalness, lip-sync accuracy, and visual quality. This is due to the difficulty of optimizing among competing preference objectives, which often conflict with one another, and the scarcity of large-scale, high-quality datasets with multidimensional preference annotations. To address these, we first introduce Talking-Critic, a multimodal reward model that learns human-aligned reward functions to quantify how well generated videos satisfy multidimensional expectations. Leveraging this model, we curate Talking-NSQ, a large-scale multidimensional human preference dataset containing 410K preference pairs. Finally, we propose Timestep-Layer adaptive multi-expert Preference Optimization (TLPO), a novel framework for aligning diffusion-based portrait animation models with fine-grained, multidimensional preferences. TLPO decouples preferences into specialized expert modules, which are then fused across timesteps and network layers, enabling comprehensive, fine-grained enhancement across all dimensions without mutual interference. Experiments demonstrate that Talking-Critic significantly outperforms existing methods in aligning with human preference ratings. Meanwhile, TLPO achieves substantial improvements over baseline models in lip-sync accuracy, motion naturalness, and visual quality, exhibiting superior performance in both qualitative and quantitative evaluations.
Random Sample Consensus (RANSAC) is a fundamental approach for robustly estimating parametric models from noisy data. Existing learning-based RANSAC methods utilize deep learning to enhance the robustness of RANSAC against outliers. However, these approaches are trained and tested on the data generated by the same algorithms, leading to limited generalization to out-of-distribution data during inference. Therefore, in this paper, we introduce a novel diffusion-based paradigm that progressively injects noise into ground-truth data, simulating the noisy conditions for training learning-based RANSAC. To enhance data diversity, we incorporate Monte Carlo sampling into the diffusion paradigm, approximating diverse data distributions by introducing different types of randomness at multiple stages. We evaluate our approach in the context of feature matching through comprehensive experiments on the ScanNet and MegaDepth datasets. The experimental results demonstrate that our Monte Carlo diffusion mechanism significantly improves the generalization ability of learning-based RANSAC. We also develop extensive ablation studies that highlight the effectiveness of key components in our framework.
Video restoration (VR) aims to recover high-quality videos from degraded ones. Although recent zero-shot VR methods using pre-trained diffusion models (DMs) show good promise, they suffer from approximation errors during reverse diffusion and insufficient temporal consistency. Moreover, dealing with 3D video data, VR is inherently computationally intensive. In this paper, we advocate viewing the reverse process in DMs as a function and present a novel Maximum a Posterior (MAP) framework that directly parameterizes video frames in the seed space of DMs, eliminating approximation errors. We also introduce strategies to promote bilevel temporal consistency: semantic consistency by leveraging clustering structures in the seed space, and pixel-level consistency by progressive warping with optical flow refinements. Extensive experiments on multiple virtual reality tasks demonstrate superior visual quality and temporal consistency achieved by our method compared to the state-of-the-art.
JoDiffusion: Jointly Diffusing Image with Pixel-Level Annotations for Semantic Segmentation Promotion
PDF ↗Given the inherently costly and time-intensive nature of pixel-level annotation, the generation of synthetic datasets comprising sufficiently diverse synthetic images paired with ground-truth pixel-level annotations has garnered increasing attention recently for training high-performance semantic segmentation models. However, existing methods necessitate to either predict pseudo annotations after image generation or generate images conditioned on manual annotation masks, which incurs image-annotation semantic inconsistency or scalability problem. To migrate both problems with one stone, we present a novel dataset generative diffusion framework for semantic segmentation, termed JoDiffusion. Firstly, given a standard latent diffusion model, JoDiffusion incorporates an independent annotation variational auto-encoder (VAE) network to map annotation masks into the latent space shared by images. Then, the diffusion model is tailored to capture the joint distribution of each image and its annotation mask conditioned on a text prompt. By doing these, JoDiffusion enables simultaneously generating paired images and semantically consistent annotation masks solely conditioned on text prompts, thereby demonstrating superior scalability. Additionally, a mask optimization strategy is developed to mitigate the annotation noise produced during generation. Experiments on Pascal VOC, COCO, and ADE20K datasets show that the annotated dataset generated by JoDiffusion yields substantial performance improvements in semantic segmentation compared to existing methods.
The introduction of diffusion models has brought significant advances to the field of audio-driven talking head generation. However, the extremely slow inference speed severely limits the practical implementation of diffusion-based talking head generation models. In this study, we propose READ, a real-time diffusion-transformer-based talking head generation framework. Our approach first learns a spatiotemporal highly compressed video latent space via a temporal VAE, significantly reducing the token count to accelerate generation. To achieve better audio-visual alignment within this compressed latent space, a pre-trained Speech Autoencoder (SpeechAE) is proposed to generate temporally compressed speech latent codes corresponding to the video latent space. These latent representations are then modeled by a carefully designed Audio-to-Video Diffusion Transformer (A2V-DiT) backbone for efficient talking head synthesis. Furthermore, to ensure temporal consistency and accelerated inference in extended generation, we propose a novel asynchronous noise scheduler (ANS) for both the training and inference processes of our framework. The ANS leverages asynchronous add-noise and asynchronous motion-guided generation in the latent space, ensuring consistency in generated video clips. Experimental results demonstrate that READ outperforms state-of-the-art methods by generating competitive talking head videos with significantly reduced runtime, achieving an optimal balance between quality and speed while maintaining robust metric stability in long-time generation.
Diffusion models have demonstrated remarkable success in various visual generation tasks, including image, video, and 3D content generation. Preference optimization (PO) is a prominent and growing area of research that aims to align these models with human preferences. While existing PO methods primarily concentrate on producing favorable outputs, they often overlook the significance of classifier-free guidance (CFG) in mitigating undesirable results. DiffusionNPO addresses this gap by introducing negative preference optimization (NPO), training models to generate outputs opposite to human preferences and thereby steering them away from unfavorable outcomes through CFG. However, prior NPO approaches rely on costly and fragile procedures for obtaining explicit preference annotations (e.g., manual pairwise labeling or reward model training), limiting their practicality in domains where such data are scarce or difficult to acquire. In this work, we propose Self-NPO, specifically truncated diffusion fine-tuning, a data-free approach of negative preference optimization by directly learning from the model itself, eliminating the need for manual data labeling or reward model training. This data-free approach is highly efficient (less than 1% training cost of Diffusion-NPO) and achieves comparable performance of Diffusion-NPO in a data-free manner. We demonstrate that Self-NPO integrates seamlessly into widely used diffusion models, including SD1.5, SDXL, and CogVideoX, as well as models already optimized for human preferences, consistently enhancing both their generation quality and alignment with human preferences.
The rapid advancement of image-generation technologies has made it possible for anyone to create photorealistic images using generative models, raising significant security concerns. To mitigate malicious use, tracing the origin of such images is essential. Reconstruction-based attribution methods offer a promising solution, but they often suffer from reduced accuracy and high computational costs when applied to state‑of‑the‑art (SOTA) models. To address these challenges, we propose AEDR (AutoEncoder Double-Reconstruction), a novel training‑free attribution method designed for generative models with continuous autoencoders. Unlike existing reconstruction‑based approaches that rely on the value of a single reconstruction loss, AEDR performs two consecutive reconstructions using the model’s autoencoder, and adopts the ratio of these two reconstruction losses as the attribution signal. This signal is further calibrated using the image homogeneity metric to improve accuracy, which inherently cancels out absolute biases caused by image complexity, with autoencoder‑based reconstruction ensuring superior computational efficiency. Experiments on eight top latent diffusion models show that AEDR achieves 25.5% higher attribution accuracy than existing reconstruction‑based methods, with requiring only 1% of the computational time.
In the context of global population aging, the prevalence of neurodegenerative diseases is rapidly increasing. Vision-based impaired gait analysis emerges as a promising alternative for automatic and non-invasive diagnosis. While prior efforts have advanced either accuracy or interpretability of gait analysis, few have effectively addressed both aspects in a unified framework. To bridge this gap, we propose DPPD, a Diffusion-based Personalized Pathology Disentanglement model that jointly performs quantitative gait scoring, dementia subtyping, and qualitative anomaly highlighting. Motivated by the observation that pathological gait features exhibit stronger inter-class separability across different gait severity than raw features, DPPD is proposed based on the subject-specific pathology disentanglement perspective. Specifically, it comprises three key components: (1) a 3DmotionBERT for encoding gait representation from 3D human pose sequences estimated, (2) a latent diffusion-based Gait Denoiser for generating personalized normal gait features, and (3) a Dual Pathology Disentanglement mechanism that captures both static pose and dynamic motion pathological representation from the residual between raw and normal gait features. These disentangled pathologies further enable quantitative classification and qualitative anomaly highlighting. Experiments on the PDGait and 3DGait datasets demonstrate that DPPD outperforms state-of-the-art methods in classification accuracy while providing reliable and interpretable visualizations of gait anomalies.
The success of diffusion models has raised concerns about the generation of unsafe or harmful content, prompting concept erasure approaches that fine-tune modules to suppress specific concepts while preserving general generative capabilities. However, as the number of erased concepts grows, these methods often become inefficient and ineffective, since each concept requires a separate set of fine-tuned parameters and may degrade the overall generation quality. In this work, we propose a supertype-subtype concept hierarchy that organizes erased concepts into a parent–child structure. Each erased concept is treated as a child node, and semantically related concepts (e.g., macaw, and bald eagle) are grouped under a shared parent node, referred to as a supertype concept (e.g., bird). Rather than erasing concepts individually, we introduce an effective and efficient group-wise suppression method, where semantically similar concepts are grouped and erased jointly by sharing a single set of learnable parameters. During the erasure phase, standard diffusion regularization is applied to preserve denoising process in unmasked regions. To mitigate the degradation of supertype generation caused by excessive erasure of semantically related subtypes, we propose a novel method called Supertype-Preserving Low-Rank Adaptation (SuPLoRA), which encodes the supertype concept information in the frozen down-projection matrix and updates only the up-projection matrix during erasure. Theoretical analysis demonstrates the effectiveness of SuPLoRA in mitigating generation performance degradation. We construct a more challenging benchmark that requires simultaneous erasure of concepts across diverse domains, including celebrities, objects, and pornographic content. Comprehensive experiments demonstrate that our method achieves a superior balance between effective multi-concept erasure and the preservation of desirable generative performance.
FedSDA: Federated Stain Distribution Alignment for Non-IID Histopathological Image Classification
PDF ↗Federated learning (FL) has shown success in collaboratively training a model among decentralized data resources without directly sharing privacy-sensitive training data. Despite recent advances, non-IID (non-independent and identically distributed) data poses an inevitable challenge that hinders the use of FL. In this work, we address the issue of non-IID histopathological images with feature distribution shifts from an intuitive perspective that has only received limited attention. Specifically, we address this issue from the perspective of data distribution by solely adjusting the data distributions of all clients. Building on the success of diffusion models in fitting data distributions and leveraging stain separation to extract the pivotal features that are closely related to the non-IID properties of histopathological images, we propose a Federated Stain Distribution Alignment (FedSDA) method. FedSDA aligns the stain distribution of each client with a target distribution in an FL framework to mitigate distribution shifts among clients. Furthermore, considering that training diffusion models on raw data in FL has been shown to be susceptible to privacy leakage risks, we circumvent this problem while still effectively achieving alignment. Extensive experimental results show that FedSDA is not only effective in improving baselines that focus on mitigating disparities across clients’ model updates but also outperforms baselines that address the non-IID data issues from the perspective of data distribution. We show that FedSDA provides valuable and practical insights for the computational pathology community.
CoGrad3D: Spatially-Coupled Timestep Optimization with Orthogonal Gradient Fusion for 3D Generation
PDF ↗Score Distillation Sampling has driven recent advances in text-to-3D generation. However, current approaches often fail to produce 3D assets that are both rich in detail and consistent across viewpoints. These limitations primarily arise from imbalanced guidance on fine-grained details and an overdependence on single-view optimization—issues exacerbated by the excessive randomness in selecting diffusion timesteps and camera configurations. Such deficiencies commonly lead to blurry textures and inter-view inconsistencies, which degrade visual realism and hinder practical deployment. To tackle these challenges, we introduce CoGrad3D, a unified generative refinement framework that adopts a continuously adaptive optimization strategy. By dynamically modulating the optimization focus based on real-time convergence signals, CoGrad3D ensures balanced progress toward both geometric completeness and high-fidelity detail. Concretely, we propose an adaptive region sampling strategy that emphasizes under-converged viewing areas, promoting stable and uniform optimization. To facilitate the transition from coarse geometry to fine-grained reconstruction, we develop a region-aware temporal scheduling scheme that integrates global training dynamics with local convergence feedback. Furthermore, we introduce a gradient fusion mechanism that consolidates historical gradients from adjacent viewpoints, mitigating view-specific artifacts and promoting the emergence of coherent 3D structures. Extensive experiments demonstrate that CoGrad3D substantially surpasses existing methods in both geometric consistency and texture fidelity, enabling the generation of high-quality, view-consistent 3D models from textual descriptions.
While diffusion models show promise for intent-based grasp generation, their isotropic noise schedules struggle with joint-specific sensitivity and task-aware variability. This limitation leads to grasps with suboptimal semantic alignment or physical feasibility. To address this challenge, we propose Semantic-guided Noise Scaling for grasp generation (SNS-Grasp), a novel framework that integrates two key innovations. First, the Semantic-guided Noise Scaling Diffusion (SNS-Diff) module generates intent-aware grasps by replacing isotropic noise with anisotropic modulation, dynamically adapting to task semantics and joint-specific sensitivity. Specifically, SNS-Diff leverages a pretrained Intent Recognizer to extract task-aware confidence scores and joint-specific gradient sensitivities from the interaction context. These signals adjust the noise scaling during denoising, downweighting perturbations for semantically critical joints to ensure semantic alignment. Second, the Fine-grained Grasp Refinement (FGR) module establishes dynamic joint-vertex coupling through fine-grained hand-object spatial relationships, enabling iterative optimization of physically executable grasps. Extensive experiments on OakInk and GRAB demonstrate SNS-Grasp's superior performance in semantic accuracy and physical feasibility, with robust generalization to unseen objects.
Since high-fidelity reference images are difficult to obtain in real underwater scenes, most deep models trained by synthetic paired data cannot match real-world data exactly. In this paper, we propose an unsupervised training framework for underwater image enhancement (UIE) by leveraging an iterative training strategy and quantification of specific neural units. Specifically, to eliminate the heavy color cast and distortion in the underwater images, we decompose the unsupervised image enhancement as two targeted sub-tasks, namely colorization and color compensation. First, a diffusion model is introduced for colorization to correct the green and blue color casts. Then, to intensify the learning ability of balanced color information, we introduce an extra network branch and propose a quantification mechanism for color compensation. The extra branch encodes style information from normal images into the generative model, while the quantification mechanism identifies and adjusts neural units relevant to warm colors, improving the model’s ability to learn balanced color feature representations for robust generation. In the end, through iterative training, color cast and distortion are progressively reduced, leading to a gradual improvement in the quality of the generated images. Experimental results on various widely used underwater datasets demonstrate that our approach achieves excellent performance, even when compared to recent supervised methods.
Diffusion models have revealed powerful potential in all-in-one image restoration (AiOIR), which is talented in generating abundant texture details. The existing AiOIR methods either retrain a diffusion model or fine-tune the pretrained diffusion model with extra conditional guidance. However, they often suffer from high inference costs and limited adaptability to diverse degradation types. In this paper, we propose an efficient AiOIR method, Diffusion Once and Done (DOD), which aims to achieve superior restoration performance with only one-step sampling of Stable Diffusion (SD) models. Specifically, multi-degradation feature modulation is first introduced to capture different degradation prompts with a pretrained diffusion model. Then, parameter-efficient conditional low-rank adaptation integrates the prompts to enable the fine-tuning of the SD model for adapting to different degradation types. Besides, a high-fidelity detail enhancement module is integrated into the decoder of SD to improve structural and textural details. Experiments demonstrate that our method outperforms existing diffusion-based restoration approaches in both visual quality and inference efficiency.
Diffusion models have emerged as state-of-the-art generative methods, particularly excelling in conditional tasks such as prompt-driven image synthesis. While recent research emphasizes the pivotal role of noise seeds in enhancing text-image alignment and generating human-preferred outputs,these works predominantly rely on random Gaussian noise or heuristic local adjustments, , overlooking the potential of global optimization trategies to systematically improve generation quality. To bridge this gap, we propose Seed Optimization based on Evolution (SOE), a hybrid framework that integrates global evolutionary search with local semantic refinement. The global evolutionary stage conducts seed selection by jointly optimizing text-image alignment (via CLIP-Score) and human preference estimation (via ImageReward), while the local stage employs diffusion inversion to inject conditional semantics into the noise seed. Together, these components constitute a model-agnostic, training-free optimization framework for conditional diffusion models. Extensive experiments across various diffusion models demonstrate that SOE consistently improves semantic fidelity and visual quality, highlighting its generalizability and potential as a plug-and-play enhancement for generative diffusion pipelines.