CVPR 2026 论文列表

Reinforcement Learning (RL) has achieved remarkable success in various domains, yet it often relies on carefully designed programmatic reward functions to guide agent behavior. Designing such reward functions can be challenging and may not generalize well across different tasks. To address this limitation, we leverage the rich world knowledge contained in pretrained video diffusion models to provide goal-driven reward signals for RL agents without ad-hoc design of reward. Our key idea is to exploit off-the-shelf video diffusion models pretrained on large-scale video datasets as informative reward functions in terms of video-level and frame-level goals. For video-level rewards, we first finetune a pretrained video diffusion model on domain-specific datasets and then employ its video encoder to evaluate the alignment between the latent representations of agent's trajectories and the generated goal videos. To enable more fine-grained goal-achievement, we derive a frame-level goal by identifying the most relevant frame from the generated video using CLIP, which serves as the goal state. We then employ a learned forward-backward representation that represents the probability of visiting the goal state from a given state-action pair as frame-level reward, promoting more coherent and goal-driven trajectories. Experiments on Meta-World and Distracting Control Suite demonstrate the effectiveness of our approach.

Training frontier-scale diffusion models often requires substantial computational resources concentrated in tightly-coupled clusters, limiting participation to well-resourced institutions. While Decentralized Diffusion Models (DDM) enable training multiple experts in isolation, existing approaches require 1176 GPU-days and homogeneous training objectives across all experts. We present an efficient framework that dramatically reduces resource requirements while supporting heterogeneous training objectives. Our approach combines three key contributions: (1) a heterogeneous decentralized training paradigm that allows experts to use different objectives (DDPM and Flow Matching), unified at inference time without any retraining; (2) pretrained checkpoint conversion from ImageNet-DDPM to Flow Matching objectives, accelerating convergence and enabling initialization without objective-specific pretraining; and (3) PixArt-\alpha's efficient AdaLN-Single architecture, reducing parameters while maintaining quality. Experiments on LAION-Aesthetics show that, relative to the training scale reported for prior DDM work, our approach reduces the compute by 16xand data by 14x. Under aligned inference settings, our heterogeneous configuration achieves better FID and higher intra-prompt diversity than the homogeneous baseline. By eliminating synchronization requirements and enabling mixed DDPM/FM objectives, our framework makes decentralized generative model training accessible to contributors with single GPUs requiring only 24--48GB VRAM.

With the rapid advancement of diffusion models, talking face generation has made remarkable progress. However, existing diffusion-based methods still require task-specific fine-tuning and large-scale audiovisual datasets, resulting in high computational costs that hinder accessibility for resource-constrained researchers and small teams. To address this, we propose a fine-tuning-free paradigm that directly performs talking face generation using the pretrained weights of Stable Diffusion and IP-Adapter. This backbone leverages the visual embedding capability of IP-Adapter to mine lip-related semantics from the pretrained Stable Diffusion. To address the challenges of identity drift, synchronization errors, and temporal instability, we also design three trainable-parameter-free components: 1) the Structurist, which explicitly disentangles and reassembles lip and appearance features to mitigate identity drift and appearance distortion; 2) the Structure Controller, which adaptively refines embeddings based on quasi-monotonic motion trends for precise lip synchronization; and 3) the Noise Sensor, which introduces a Gaussian prior to detect and suppress flicker and jitter artifacts and enhance temporal consistency. Experimental results show that our method outperforms existing SOTA approaches in both lip-sync accuracy (at least 0.16 gain in PCLD) and visual fidelity (at least 0.7 improvement in FID), establishing a novel fine-tuning-free diffusion framework for talking face generation.

Recent advancements in 3D generative modeling have significantly improved the generation realism, yet the field is still hampered by existing representations, which struggle to capture assets with complex topologies and detailed appearance. This paper present an approach for learning a structured latent representation from native 3D data to address this challenge. At its core is a new sparse voxel structure called O-Voxel, an omni-voxel representation that encodes both geometry and appearance. O-Voxel can robustly model arbitrary topology, including open, non-manifold, and fully-enclosed surfaces, while capturing comprehensive surface attributes beyond texture color, such as physically-based rendering parameters. Based on O-Voxel, we design a Sparse Compression VAE which provides a high spatial compression rate and a compact latent space. We train large-scale flow-matching models comprising 4B parameters for 3D generation using diverse public 3D asset datasets. Despite their scale, inference remains highly efficient. Meanwhile, the geometry and material quality of our generated assets far exceed those of existing models. We believe our approach offers a significant advancement in 3D generative modeling.

Text-guided diffusion models have advanced image editing by enabling intuitive control through language. However, despite their strong capabilities, we surprisingly find that SOTA methods struggle with simple, everyday transformations such as rain or blur. We attribute this limitation to weak and inconsistent textual supervision during training, which leads to poor alignment between language and vision. Existing solutions often rely on extra finetuning or stronger text conditioning, but suffer from high data and computational requirements. We argue that diffusion-based editing capabilities aren't lost but merely hidden from text. The door to cost-efficient visual editing remains open, and the key lies in a vision-centric paradigm that perceives and reasons about visual change as humans do, beyond words. Inspired by this, we introduce Visual Diffusion Conditioning (VDC), a training-free framework that learns conditioning signals directly from visual examples for precise, language-free image editing. Given a paired example--one image with and one without the target effect--VDC derives a visual condition that captures the transformation and steers generation through a novel condition-steering mechanism. An accompanying inversion-correction step mitigates reconstruction errors during DDIM inversion, preserving fine detail and realism. Across diverse tasks, VDC outperforms both training-free and fully fine-tuned text-based editing methods. The code and models are open-sourced at https://omaralezaby.github.io/vdc/

Distribution Matching Distillation (DMD) distills score-based generative models into efficient one-step generators, without requiring a one-to-one correspondence with the sampling trajectories of their teachers. Yet, the limited capacity of one-step distilled models compromises generative diversity and degrades performance in complex generative tasks, e.g., generating intricate object motions in text-to-video task. Directly extending DMD to multi-step distillation increases memory usage and computational depth, leading to instability and reduced efficiency. While prior works propose stochastic gradient truncation as a potential solution, we observe that it substantially reduces the generative diversity in text-to-image generation and slows motion dynamics in video generation, reducing performance to the level of one-step models. To address these limitations, we propose Phased DMD, a multi-step distillation framework that bridges the idea of phase-wise distillation with Mixture-of-Experts (MoE), reducing learning difficulty while enhancing model capacity. Phased DMD incorporates two key ideas: progressive distribution matching and score matching within subintervals. First, our model divides the SNR range into subintervals, progressively refining the model to higher SNR levels, to better capture complex distributions. Next, to ensure accurate training within each subinterval, we derive rigorous mathematical formulations for the objective. We validate Phased DMD by distilling state-of-the-art image and video generation models, including Qwen-Image-20B and Wan2.2-28B. Experiments demonstrate that Phased DMD enhances motion dynamics, improves visual fidelity in video generation, and increases output diversity in image generation. Our code and models are available at https://x-niper.github.io/projects/Phased-DMD/.

Single-view 3D human reconstruction has garnered significant attention in recent years. Despite numerous advancements, prior research has concentrated on reconstructing 3D models from clear, close-up images of individual subjects, often yielding subpar results in the more prevalent multi-person scenarios. Reconstructing 3D human crowd models is a highly intricate task, laden with challenges such as: 1) extensive occlusions, 2) low clarity, and 3) numerous and various appearances. To address this task, we propose CrowdGaussian, a unified framework that directly reconstructs multi-person 3D Gaussian Splatting (3DGS) representations from single-image inputs. To handle occlusions, we devise a self-supervised adaptation pipeline that enables the pretrained large human model to reconstruct complete 3D humans with plausible geometry and appearance from heavily occluded inputs. Furthermore, we introduce Self-Calibrated Learning (SCL). This training strategy enables single-step diffusion models to adaptively refine coarse renderings to optimal quality by blending identity-preserving samples with clean/corrupted image pairs. The outputs can be distilled back to enhance the quality of multi-person 3DGS representations. Extensive experiments demonstrate that CrowdGaussian generates photorealistic, geometrically coherent reconstructions of multi-person scenes.

Existing feed-forward networks excel at predicting a single set of physical properties from visual appearance, but this point-estimate paradigm fundamentally fails to capture the real world's inherent physical ambiguity. We address this by reframing physics prediction as a task of learning a controllable, continuous distribution of material properties. We introduce UNIPIXIE, a framework trained to predict a continuous and parameterized path of physically plausible material properties from a single visual input. By learning a direct mapping along an object's softest-to-stiffest spectrum on our PIXIEMULTIVERSE dataset, UNIPIXIE allows for controllable generation of diverse, physically-valid material fields via a single intuitive parameter. Crucially, UNIPIXIE introduces a novel unified architecture to produce simulation-ready parameters for diverse physics solvers, including continuum-based Material Point Method (MPM), reduced-order deformation based on Linear Blend Skinning (LBS), and anchor-based Spring-Mass systems, addressing a key portability issue in prior work. Experiments show our approach not only generates a rich variety of plausible dynamics but also reduces Young's Modulus prediction error by over 50% against the strongest deterministic baseline, bridging the gap between static point-estimates and the continuous nature of physical reality.

Simulation of cellular morphology change has long been a fundamental task in quantitative biology and high-throughput screening, with the potential to accelerate therapeutic development and elucidate disease mechanisms beyond empirical clinical practice. However, the vast perturbation space poses challenges to the discriminative formulation, and existing generative approaches tend to concentrate on the same trajectory subspace, making their generated paths prone to drift. In this paper, we propose a novel Steered Diffusion Bridge approach, named SimuSDB, to define deterministic probabilistic trajectories between two distinct state domains for cell response generation. We first extend the diffusion bridge paradigm to maintain stochasticity and diversity in interpolation trajectories by introducing Brownian bridges. Then, SimuSDB generates cell morphologies that comply with phenotypic constraints, while allowing the latter to explicitly guide the generative process. For the inference stage, we formalize the rule-guided sample generation task as an optimal control problem within a stochastic dynamical system. This way, the generative model can achieve analytically tractable optimal control strategies and steered generation without collapsing toward the trajectory of the same data subspace. Comprehensive experiments demonstrate the superior performance of SimuSDB across various applications, including chemical perturbation and genetic perturbation.

The Mean Flow Matching algorithm is the state-of-the-art for one-step generative models. Building on this idea, we propose the Stable Mean Flow algorithm and introduce a Lyapunov-inspired stability regularizer that enforces local non-expansivity of the single-step transport map. This design guarantees uniqueness of characteristics and bounds trajectory drift. We conduct experiments that show improved output quality and convergence speed over Mean Flow. Moreover, we establish explicit upper bounds on error growth for both one-step and multi-step generation.

Our goal is to train a generative model of 3D hand motions, conditioned on natural language descriptions specifying motion characteristics such as handshapes, locations, finger/hand/arm movements. To this end, we automatically build pairs of 3D hand motions and their associated textual labels with unprecedented scale. Specifically, we leverage a large-scale sign language video dataset, along with noisy pseudo-annotated sign categories, which we translate into hand motion descriptions via an LLM that utilizes a dictionary of sign attributes, as well as our complementary motion-script cues. This data enables training a text-conditioned hand motion diffusion model (HandMDM), that is robust across domains such as unseen sign categories from the same sign language, but also signs from another sign language and non-sign hand movements. We contribute extensive experimental investigation of these scenarios and will make our trained models and data publicly available to support future research in this relatively new field.

Chain-of-Thought (CoT) has recently shown encouraging progress in the vision language model. However, the pure-vision CoT (i.e., chain-of-vision) has been underexplored in visual in-context learning. In this paper, we introduce Diffusion Guided Chain-of-Vision, which integrates an explicit chain-of-thought process into autoregressive vision models through vision prior from pre-trained diffusion models. Concretely, we find that pre-trained diffusion models induce a reliable probability flow in image space, where intermediate images sampled along this flow exhibit visual coherence and serve as task-free, chain-of-vision supervision for pure-vision autoregressive models. Extensive experiments on diverse vision tasks and multi-scale models validate the effectiveness of our proposed method for visual in-context learning. Code and dataset will be publicly available.

We present Flowception, a novel non-autoregressive and variable-length video generation framework. Flowception learns a probability path that interleaves discrete frame insertions with continuous frame denoising. Compared to autoregressive methods, Flowception alleviates error accumulation/drift as the frame insertion mechanism during sampling serves as an efficient compression mechanism to handle long-term context. Compared to full-sequence flows, our method reduces FLOPs for training three-fold, while also being more amenable to local attention variants, and allowing to learn the length of videos jointly with their content. Quantitative experimental results show improved FVD and VBench metrics over autoregressive and full-sequence baselines, which is further validated with qualitative results. Finally, by learning to insert and denoise frames in a sequence, Flowception seamlessly integrates different tasks such as image-to-video generation and video interpolation.

The performance of Latent Diffusion Models (LDMs) is critically dependent on the quality of their visual tokenizers. While recent works have explored incorporating Vision Foundation Models (VFMs) into the tokenizers training via distillation, we empirically find this approach inevitably weakens the robustness of learnt representation from original VFM. In this paper, we bypass the distillation by proposing a more direct approach by leveraging the frozen VFM for the LDMs tokenizer, named VFM Variational Autoencoder (VFM-VAE). To fully exploit the potential to leverage frozen VFM for the LDMs tokenizer, we design a new decoder to reconstruct realistic images from the semantic-rich representation of VFM. With the proposed VFM-VAE, we conduct a systematic study on how the representation from different tokenizers impact the representation learning process throughout diffusion training, enabling synergistic benefits of dual-side alignment on both tokenizers and diffusion models. Our effort in tokenizer design and training strategy lead to superior performance and efficiency: our system reaches a **gFID (w/o CFG) of 2.22 in merely 80 epochs** (a 10xspeedup over prior tokenizers). With continued training to **640 epochs**, it further attains a **gFID (w/o CFG) of 1.62**. These results offer solid evidence for the substantial potential of VFMs to serve as visual tokenizers to accelerate the LDM training progress.

Recent advancements in text-to-image synthesis have been largely propelled by diffusion-based models, yet achieving precise alignment between text prompts and generated images remains a persistent challenge. We find that this difficulty arises primarily from the limitations of conventional diffusion loss, which provides only implicit supervision for modeling fine-grained text-image correspondence. In this paper, we introduce Cross-Timestep Self-Calibration (CTCal), founded on the supporting observation that establishing accurate text-image alignment within diffusion models becomes progressively more difficult as the timestep increases. CTCal leverages the reliable text-image alignment (i.e., cross-attention maps) formed at smaller timesteps with less noise to calibrate the representation learning at larger timesteps with more noise, thereby providing explicit supervision during training. We further propose a timestep-aware adaptive weighting to achieve a harmonious integration of CTCal and diffusion loss. CTCal is model-agnostic and can be seamlessly integrated into existing text-to-image diffusion models, encompassing both diffusion-based (e.g., SD 2.1) and flow-based approaches (e.g., SD 3). Extensive experiments on T2I-Compbench++ and GenEval benchmarks demonstrate the effectiveness and generalizability of the proposed CTCal. Our code is available at https://github.com/xiefan-guo/ctcal.

The rapid evolution of deepfake technologies demands robust and reliable face forgery detection algorithms. While determining whether an image has been manipulated remains essential, the ability to precisely localize forgery clues is also important for enhancing model explainability and building user trust. To address this dual challenge, we introduce DiffusionFF, a diffusion-based framework that simultaneously performs face forgery detection and fine-grained artifact localization. Our key idea is to establish a novel encoder-decoder architecture: a pretrained forgery detector serves as a powerful "artifact encoder", and a denoising diffusion model is repurposed as an "artifact decoder". Conditioned on multi-scale forgery-related features extracted by the encoder, the decoder progressively synthesizes a detailed artifact localization map. We then fuse this fine-grained localization map with high-level semantic features from the forgery detector, leading to substantial improvements in detection capability. Extensive experiments show that DiffusionFF achieves state-of-the-art (SOTA) performance across multiple benchmarks, underscoring its superior effectiveness and explainability.

Existing 3D editing methods rely on computationally intensive scene-by-scene iterative optimization and suffer from multi-view inconsistency. We propose an effective and feed-forward 3D editing framework based on the TRELLIS generative backbone, capable of modifying 3D models from a single editing view. Our framework addresses two key issues: adapting training-free 2D editing to structured 3D representations, and overcoming the bottleneck of appearance fidelity in compressed 3D features. To ensure geometric consistency, we introduce Voxel FlowEdit, an edit-driven flow in the sparse voxel latent space that achieves globally consistent 3D deformation in a single pass. To restore high-fidelity details, we develop a normal-guided single to multi-view generation module as an external appearance prior, successfully recovering high-frequency textures. Experiments demonstrate that our method enables fast, globally consistent, and high-fidelity 3D model editing.

We propose Mesh4D, a feed-forward model for monocular 4D mesh reconstruction. Given a monocular video of a dynamic object, our model reconstructs the object's complete 3D shape and motion, represented as a deformation field. Our key contribution is a compact latent space that encodes the entire animation sequence. This latent space is learned by an autoencoder that, during training, is guided by the skeletal structure of the training objects, providing strong priors on plausible deformations. Crucially, skeletal information is not required at inference time. The encoder employs spatio-temporal attention, yielding a more stable representation of the object's overall deformation. Building on this representation, we train a latent diffusion model that, conditioned on the input video and the mesh reconstructed from the first frame, predicts the full animation in one shot. We evaluate Mesh4D on reconstruction and novel view synthesis benchmarks, outperforming prior methods in recovering accurate 3D shape and deformation.

Efficient and robust feature matching is crucial for latency-sensitive and resource-constrained applications. However, many current semi-dense feature matching methods scale quadratically with spatial resolution, either due to transformer-based long-range context modeling or from redundant full correlation computations. To overcome these limitations, we present SLiM (Salient Lightweight Matching), a novel scalable feature matching method that delivers reliable matches with low memory footprint and latency, especially at high resolutions. Our approach introduces three key innovations: (1) a hybrid Conv-Mamba backbone for efficient cross-scale and cross-view feature extraction with linear complexity, (2) a training-free norm-based feature filtering mechanism, enabling sparse correlation that significantly reduces computation overhead during inference, and (3) a lightweight recurrent coordinate refinement that surpasses expectation-based regression in subpixel accuracy. Experimental results show that SLiM consistently achieves a strong accuracy-efficiency trade-off across indoor and outdoor benchmarks, with clear advantages in memory usage, inference speed, and high-resolution scalability. These results demonstrate the practical efficiency and strong scalability of SLiM. Project page: https://github.com/Band-127/SLiM