Diffusion Models

Follow this topic Ask about this topic

Diffusion models learn a data distribution by gradually corrupting examples with noise and training a denoiser to reverse that process. Major approaches include DDPMs and score-based SDEs, latent diffusion and diffusion Transformers, flow and consistency methods, and discrete masked or continuous-time Markov diffusion; conditioning and guidance adapt them to specific tasks. They are used for images, video, audio, language, molecules, and scientific forecasting, with evaluations ranging from ImageNet FID to language perplexity and domain-specific measures of accuracy and uncertainty. A central tension is how to make sampling faster and generation more controllable without sacrificing quality, calibration, or diversity, while theory and ongoing audits address generalization, memorization, and privacy.

Latest papers

1,192 papers152 papers in the last four weeks, up 77% on the four weeks before. 0.9% of all new papers.

Apr 19, 2026

Interpolating Discrete Diffusion Models with Controllable Resampling

Discrete diffusion models form a powerful class of generative models across diverse domains, including text and graphs. However, existing approaches face fundamental limitations. Masked diffusion models suffer from irreversible errors due to early unmasking, while uniform diffusion models, despite enabling self-correction, often yield low-quality samples due to their strong reliance on intermediate latent states. We introduce IDDM, an Interpolating Discrete Diffusion Model, that improves diffusion by reducing dependence on intermediate latent states. Central to IDDM is a controllable resampling mechanism that partially resets probability mass to the marginal distribution, mitigating error accumulation and enabling more effective token corrections. IDDM specifies a generative process whose transitions interpolate between staying at the current state, resampling from a prior, and flipping toward the target state, while enforcing marginal consistency and fully decoupling training from inference. We benchmark our model against state-of-the-art discrete diffusion models across molecular graph generation as well as text generation tasks, demonstrating competitive performance.

Marcel Kollovieh, Sirine Ayadi, Stephan Günnemann

School of Computation, Information and Technology, Technical University of Munich

Diffusion Model Sampling Molecular Graph Generation Text Generation Diffusion Models

Apr 18, 2026

Marrying Text-to-Motion Generation with Skeleton-Based Action Recognition

Human action recognition and motion generation are two active research problems in human-centric computer vision, both aiming to align motion with textual semantics. However, most existing works study these two problems separately, without uncovering the links between them, namely that motion generation requires semantic comprehension. This work investigates unified action recognition and motion generation by leveraging skeleton coordinates for both motion understanding and generation. We propose Coordinates-based Autoregressive Motion Diffusion (CoAMD), which synthesizes motion in a coarse-to-fine manner. As a core component of CoAMD, we design a Multi-modal Action Recognizer (MAR) that provides gradient-based semantic guidance for motion generation. Furthermore, we establish a rigorous benchmark by evaluating baselines on absolute coordinates. Our model can be applied to four important tasks, including skeleton-based action recognition, text-to-motion generation, text-motion retrieval, and motion editing. Extensive experiments on 13 benchmarks across these tasks demonstrate that our approach achieves state-of-the-art performance, highlighting its effectiveness and versatility for human motion modeling. Code is available at https://github.com/jidongkuang/CoAMD.

Jidong Kuang, Hongsong Wang, Jie Gui

School of Cyber Science and Engineering, Southeast University, Nanjing, China · School of Computer Science and Engineering, Southeast University, Nanjing, China

Skeleton-Based Action Recognition Human Activity Recognition Text-to-Motion Generation Motion Diffusion Models Human Motion Generation Diffusion Models

Apr 17, 2026

Repurposing 3D Generative Model for Autoregressive Layout Generation

We introduce LaviGen, a framework that repurposes 3D generative models for 3D layout generation. Unlike previous methods that infer object layouts from textual descriptions, LaviGen operates directly in the native 3D space, formulating layout generation as an autoregressive process that explicitly models geometric relations and physical constraints among objects, producing coherent and physically plausible 3D scenes. To further enhance this process, we propose an adapted 3D diffusion model that integrates scene, object, and instruction information and employs a dual-guidance self-rollout distillation mechanism to improve efficiency and spatial accuracy. Extensive experiments on the LayoutVLM benchmark show LaviGen achieves superior 3D layout generation performance, with 19% higher physical plausibility than the state of the art and 65% faster computation. Our code is publicly available at https://github.com/fenghora/LaviGen.

Haoran Feng, Yifan Niu, Zehuan Huang +4

School of Software, Beihang University · 2Tsinghua University · University of Hong Kong +2

3D Scene Generation Autoregressive Generation 3D Scene Layout Generation Diffusion Models Autoregressive 3D Scene Generation

Apr 17, 2026

Motion-Adapter: A Diffusion Model Adapter for Text-to-Motion Generation of Compound Actions

Recent advances in generative motion synthesis have enabled the production of realistic human motions from diverse input modalities. However, synthesizing compound actions from texts, which integrate multiple concurrent actions into coherent full-body sequences, remains a major challenge. We identify two key limitations in current text-to-motion diffusion models: (i) catastrophic neglect, where earlier actions are overwritten by later ones due to improper handling of temporal information, and (ii) attention collapse, which arises from excessive feature fusion in cross-attention mechanisms. As a result, existing approaches often depend on overly detailed textual descriptions (e.g., raising right hand), explicit body-part specifications (e.g., editing the upper body), or the use of large language models (LLMs) for body-part interpretation. These strategies lead to deficient semantic representations of physical structures and kinematic mechanisms, limiting the ability to incorporate natural behaviors such as greeting while walking. To address these issues, we propose the Motion-Adapter, a plug-and-play module that guides text-to-motion diffusion models in generating compound actions by computing decoupled cross-attention maps, which serve as structural masks during the denoising process. Extensive experiments demonstrate that our method consistently produces more faithful and coherent compound motions across diverse textual prompts, surpassing state-of-the-art approaches.

Yue Jiang, Mingyu Yang, Liuyuxin Yang +3

School of Computer Science, Northwest University, Xi’an, China

Diffusion Model Guidance Text-to-Motion Generation Human Motion Generation Diffusion Models

Apr 17, 2026

Elucidating the SNR-t Bias of Diffusion Probabilistic Models

Diffusion Probabilistic Models have demonstrated remarkable performance across a wide range of generative tasks. However, we have observed that these models often suffer from a Signal-to-Noise Ratio-timestep (SNR-t) bias. This bias refers to the misalignment between the SNR of the denoising sample and its corresponding timestep during the inference phase. Specifically, during training, the SNR of a sample is strictly coupled with its timestep. However, this correspondence is disrupted during inference, leading to error accumulation and impairing the generation quality. We provide comprehensive empirical evidence and theoretical analysis to substantiate this phenomenon and propose a simple yet effective differential correction method to mitigate the SNR-t bias. Recognizing that diffusion models typically reconstruct low-frequency components before focusing on high-frequency details during the reverse denoising process, we decompose samples into various frequency components and apply differential correction to each component individually. Extensive experiments show that our approach significantly improves the generation quality of various diffusion models (IDDPM, ADM, DDIM, A-DPM, EA-DPM, EDM, PFGM++, and FLUX) on datasets of various resolutions with negligible computational overhead. The code is at https://github.com/AMAP-ML/DCW.

Meng Yu, Lei Sun, Jianhao Zeng +2

Lanzhou University · AMAP Alibaba Group

Diffusion Model Sampling Diffusion Sampling Diffusion-Based Image Generation Diffusion Models

Apr 17, 2026

From Competition to Coopetition: Coopetitive Training-Free Image Editing Based on Text Guidance

Text-guided image editing, a pivotal task in modern multimedia content creation, has seen remarkable progress with training-free methods that eliminate the need for additional optimization. Despite recent progress, existing methods are typically constrained by a competitive paradigm in which the editing and reconstruction branches are independently driven by their respective objectives to maximize alignment with target and source prompts. The adversarial strategy causes semantic conflicts and unpredictable outcomes due to the lack of coordination between branches. To overcome these issues, we propose Coopetitive Training-Free Image Editing (CoEdit), a novel zero-shot framework that transforms attention control from competition to coopetitive negotiation, achieving editing harmony across spatial and temporal dimensions. Spatially, CoEdit introduces Dual-Entropy Attention Manipulation, which quantifies directional entropic interactions between branches to reformulate attention control as a harmony-maximization problem, eventually improving the localization of editable and preservable regions. Temporally, we present Entropic Latent Refinement mechanism to dynamically adjust latent representations over time, minimizing accumulated editing errors and ensuring consistent semantic transitions throughout the denoising trajectory. Additionally, we propose the Fidelity-Constrained Editing Score, a composite metric that jointly evaluates semantic editing and background fidelity. Extensive experiments on standard benchmarks demonstrate that CoEdit achieves superior performance in both editing quality and structural preservation, enhancing multimedia information utilization by enabling more effective interaction between visual and textual modalities. The code will be available at https://github.com/JinhaoShen/CoEdit.

Jinhao Shen, Haoqian Du, Xulu Zhang +2

Department of Computing, Hong Kong Polytechnic University, Hong Kong, China · Department of Computer Science, Sichuan University, Chengdu 610017, China

Text-Guided Image Editing Attention Control in Diffusion Models Training-Free Image Editing Diffusion-Based Image Editing Image Editing Diffusion Models

Apr 17, 2026

Co-generation of Layout and Shape from Text via Autoregressive 3D Diffusion

Recent text-to-scene generation approaches largely reduced the manual efforts required to create 3D scenes. However, their focus is either to generate a scene layout or to generate objects, and few generate both. The generated scene layout is often simple even with LLM's help. Moreover, the generated scene is often inconsistent with the text input that contains non-trivial descriptions of the shape, appearance, and spatial arrangement of the objects. We present a new paradigm of sequential text-to-scene generation and propose a novel generative model for interactive scene creation. At the core is a 3D Autoregressive Diffusion model 3D-ARD+, which unifies the autoregressive generation over a multimodal token sequence and diffusion generation of next-object 3D latents. To generate the next object, the model uses one autoregressive step to generate the coarse-grained 3D latents in the scene space, conditioned on both the current seen text instructions and already synthesized 3D scene. It then uses a second step to generate the 3D latents in the smaller object space, which can be decoded into fine-grained object geometry and appearance. We curate a large dataset of 230K indoor scenes with paired text instructions for training. We evaluate 7B 3D-ARD+, on challenging scenes, and showcase the model can generate and place objects following non-trivial spatial layout and semantics prescribed by the text instructions.

Zhenggang Tang, Yuehao Wang, Yuchen Fan +9

Meta Reality Labs · University of Illinois Urbana-Champaign · University of Texas at Austin

Autoregressive Diffusion 3D Scene Generation 3D Scene Layout Generation Indoor Scene Generation Diffusion Models Autoregressive 3D Scene Generation

Apr 17, 2026

Diffusion Autoencoder for Unsupervised Artifact Restoration in Handheld Fundus Images

The advent of handheld fundus imaging devices has made ophthalmologic diagnosis and disease screening more accessible, efficient, and cost-effective. However, images captured from these setups often suffer from artifacts such as flash reflections, exposure variations, and motion-induced blur, which degrade image quality and hinder downstream analysis. While generative models have been effective in image restoration, most depend on paired supervision or predefined artifact structures, making them less adaptable to unstructured degradations commonly observed in handheld fundus images. To address this, we propose an unsupervised diffusion autoencoder that integrates a context encoder with the denoising process to learn semantically meaningful representations for artifact restoration. The model is trained only on high-quality table-top fundus images and infers to restore artifact-affected handheld acquisitions. We validate the restorations through quantitative and qualitative evaluations, and have shown that diagnostic accuracy increases to 81.17% on an unseen dataset and multiple artifact conditions

Mathumetha Palani, Kavya Puthumana, Ayantika Das +1

Nuvo AI, Chennai, India · Indian Institute of Technology Madras, India

Diffusion Models for Image Restoration Unsupervised Image Restoration Medical Imaging Image Restoration Fundus Imaging Diffusion Autoencoder+1

Apr 17, 2026

Neural Continuous-Time Markov Chain: Discrete Diffusion via Decoupled Jump Timing and Direction

Discrete diffusion models based on continuous-time Markov chains (CTMCs) have shown strong performance on language and discrete data generation, yet existing approaches typically parameterize the reverse rate matrix monolithically -- through proxies such as concrete scores (SEDD) or clean-data predictions (MDLM, GIDD) -- rather than aligning the parameterization with the intrinsic CTMC decomposition into jump timing and jump direction. We propose Neural CTMC, which exploits the underlying Poisson structure of CTMC dynamics by separately parameterizing the reverse process through an exit rate (when to jump) and a jump distribution (where to jump) via two dedicated network heads. We show that the evidence lower bound (ELBO) reduces to a path-space KL divergence between the true and learned reverse processes that factorizes into a Poisson KL for timing and a categorical KL for direction, and admits a tractable, gradient-equivalent and consistent loss. Experimentally, scored by Gemma2-9B, our pure-uniform Neural CTMC achieves

16.36

generative perplexity on TinyStories (vs.\ GIDD

37.60

and MDLM

42.66

). On OpenWebText, it attains the best perplexity at the same training-token budget across 16--128 sampling steps among the methods we compare (e.g., at 128 steps: Neural CTMC

183.6

vs.\ MDLM

210.5

and GIDD

249.8

). To facilitate reproducibility, we release our pretrained weights at https://huggingface.co/Jiangxy1117/Neural-CTMC.

Jingyuan Li, Xiaoyi Jiang, Fukang Wen +5

2Beijing Institute of Mathematical Sciences and Applications · 3Wuhan University · 4MathonAI +1

Markov Models Diffusion Language Models Conditional Generative Modeling Diffusion Models Markov Chains

Apr 17, 2026

CLIMB: Controllable Longitudinal Brain Image Generation using Mamba-based Latent Diffusion Model and Gaussian-aligned Autoencoder

Latent diffusion models have emerged as powerful generative models in medical imaging, enabling the synthesis of high quality brain magnetic resonance imaging scans. In particular, predicting the evolution of a patients brain can aid in early intervention, prognosis, and treatment planning. In this study, we introduce CLIMB, Controllable Longitudinal brain Image generation via state space based latent diffusion model, an advanced framework for modeling temporal changes in brain structure. CLIMB is designed to model the structural evolution of the brain structure over time, utilizing a baseline MRI scan and its acquisition age as foundational inputs. Additionally, multiple conditional variables, including projected age, gender, disease status, genetic information, and brain structure volumes, are incorporated to enhance the temporal modeling of anatomical changes. Unlike existing LDM methods that rely on self attention modules, which effectively capture contextual information from input images but are computationally expensive, our approach leverages state space, a state space model architecture that substantially reduces computational overhead while preserving high-quality image synthesis. Furthermore, we introduce a Gaussian-aligned autoencoder that extracts latent representations conforming to prior distributions without the sampling noise inherent in conventional variational autoencoders. We train and evaluate our proposed model on the Alzheimers Disease Neuroimaging Initiative dataset, consisting of 6,306 MRI scans from 1,390 participants. By comparing generated images with real MRI scans, CLIMB achieves a structural similarity index of 0.9433, demonstrating notable improvements over existing methods.

Duy-Phuong Dao, Muhammad Taqiyuddin, Jahae Kim +4

Department of Artificial Intelligence Convergence, Chonnam National University, South Korea · Department of Nuclear Medicine, Chonnam National University, South Korea · University of South Australia, Australia +2

Latent Diffusion Models Longitudinal Medical Image Analysis Medical Image Synthesis Neuroimaging Brain MRI Diffusion Models

Apr 16, 2026

An Analysis of Regularization and Fokker-Planck Residuals in Diffusion Models for Image Generation

Recent work has shown that diffusion models trained with the denoising score matching (DSM) objective often violate the Fokker--Planck (FP) equation that governs the evolution of the true data density. Directly penalizing these deviations in the objective function reduces their magnitude but introduces a significant computational overhead. It is also observed that enforcing strict adherence to the FP equation does not necessarily lead to improvements in the quality of the generated samples, as often the best results are obtained with weaker FP regularization. In this paper, we investigate whether simpler penalty terms can provide similar benefits. We empirically analyze several lightweight regularizers, study their effect on FP residuals and generation quality, and show that the benefits of FP regularization are available at substantially lower computational cost. Our code is available at https://github.com/OnnoNiemann/fp_diffusion_analysis.

Onno Niemann, Gonzalo Martínez Muñoz, Alberto Suárez Gonzalez

Universidad Autónoma de Madrid

Diffusion-Based Image Generation Score-Based Generative Modeling Diffusion Models Fokker–Planck Equation

Apr 16, 2026

Diffusion Crossover: Defining Evolutionary Recombination in Diffusion Models via Noise Sequence Interpolation

Interactive Evolutionary Computation (IEC) provides a powerful framework for optimizing subjective criteria such as human preferences and aesthetics, yet it suffers from a fundamental limitation: in high-dimensional generative representations, defining crossover in a semantically consistent manner is difficult, often leading to a mutation-dominated search. In this work, we explicitly define crossover in diffusion models. We propose Diffusion crossover, which formulates evolutionary recombination as step-wise interpolation of noise sequences in the reverse process of Denoising Diffusion Probabilistic Models (DDPMs). By applying spherical linear interpolation (Slerp) to the noise sequences associated with selected parent images, the proposed method generates offspring that inherit characteristics from both parents while preserving the geometric structure of the diffusion process. Furthermore, controlling the time-step range of interpolation enables a principled trade-off between diversity (exploration) and convergence (exploitation). Experimental results using PCA analysis and perceptual similarity metrics (LPIPS) demonstrate that Diffusion crossover produces perceptually smooth and semantically consistent transitions between parent images. Qualitative interactive evolution experiments further confirm that the proposed method effectively supports human-in-the-loop image exploration. These findings suggest a new perspective: diffusion models are not only powerful generators, but also structured evolutionary search spaces in which recombination can be explicitly defined and controlled.

Chisato Kumada, Satoru Hiwa, Tomoyuki Hiroyasu

Graduate School of Life and Medical Sciences · Doshisha University · Kyoto, Japan +1

Image Generation Diffusion Models

Apr 16, 2026

SynHAT: A Two-stage Coarse-to-Fine Diffusion Framework for Synthesizing Human Activity Traces

Human activity traces (HATs) are critical for many applications, including human mobility modeling and point-of-interest (POI) recommendation. However, growing privacy concerns have severely limited access to authentic large-scale HAT datasets. Recent advances in generative AI provide new opportunities to synthesize realistic and privacy-preserving HATs for such applications. Yet two major challenges remain: (i) HATs are highly irregular and dynamic, with long and varying time intervals, making it difficult to capture their complex spatio-temporal dependencies and underlying distributions; and (ii) generative models are often computationally expensive, making long-term, fine-grained HAT synthesis inefficient. To address these challenges, we propose SynHAT, a computationally efficient coarse-to-fine HAT synthesis framework built on a novel spatio-temporal denoising diffusion model. In Stage 1, we develop Coarse-HADiff, which models the overall spatio-temporal dependencies of coarse-grained latent spatio-temporal traces. It incorporates a novel Latent Spatio-Temporal U-Net with dual Drift-Jitter branches to jointly model smooth spatial transitions and temporal variations during denoising. In Stage 2, we introduce a three-step pipeline consisting of Behavior Pattern Extraction, Fine-HADiff, which shares the same architecture as Coarse-HADiff, and Semantic Alignment to generate fine-grained latent spatio-temporal traces from the Stage 1 outputs. We extensively evaluate SynHAT in terms of data fidelity, utility, privacy, robustness, and scalability. Experiments on real-world HAT datasets from four cities across three countries show that SynHAT substantially outperforms state-of-the-art baselines, achieving 52% and 33% improvements on spatial and temporal metrics, respectively.

Rongchao Xu, Lin Jiang, Dahai Yu +2

Florida State University, USA

Synthetic Data Generation Diffusion Models

Apr 16, 2026

DVFace: Spatio-Temporal Dual-Prior Diffusion for Video Face Restoration

Video face restoration aims to enhance degraded face videos into high-quality results with realistic facial details, stable identity, and temporal coherence. Recent diffusion-based methods have brought strong generative priors to restoration and enabled more realistic detail synthesis. However, existing approaches for face videos still rely heavily on generic diffusion priors and multi-step sampling, which limit both facial adaptation and inference efficiency. These limitations motivate the use of one-step diffusion for video face restoration, yet achieving faithful facial recovery alongside temporally stable outputs remains challenging. In this paper, we propose, DVFace, a one-step diffusion framework for real-world video face restoration. Specifically, we introduce a spatio-temporal dual-codebook design to extract complementary spatial and temporal facial priors from degraded videos. We further propose an asymmetric spatio-temporal fusion module to inject these priors into the diffusion backbone according to their distinct roles. Evaluation on various benchmarks shows that DVFace delivers superior restoration quality, temporal consistency, and identity preservation compared to recent methods. Code: https://github.com/zhengchen1999/DVFace.

Zheng Chen, Bowen Chai, Rongjun Gao +7

Shanghai Jiaotong University · Meituan Inc

Video Diffusion Models Image Restoration Video Restoration Diffusion Models

Apr 15, 2026

Thermodynamic Diffusion Inference with Minimal Digital Conditioning

Diffusion-model inference and overdamped Langevin dynamics are formally identical. A physical substrate that encodes the score function therefore equilibrates to the correct output by thermodynamics alone, requiring no digital arithmetic during inference and potentially achieving a

10{,}000\times

reduction in energy relative to a GPU. Two fundamental barriers have until now prevented this equivalence from being realized at production scale: non-local skip connections, which locally coupled analog substrates cannot represent, and input conditioning, in which the coupling constants carry roughly

2{,}600\times

too little signal to anchor the system to a specific input. We resolve both obstacles. Hierarchical bilinear coupling encodes U-Net skip connections as rank-

k

inter-module interactions derived directly from the singular structure of the encoder and decoder Gram matrices, requiring only

O(Dk)

physical connections instead of

O(D^2)

. A minimal digital interface -- a 4-dimensional bottleneck encoder together with a 16-unit transfer network, totalling 2,560 parameters -- overcomes the conditioning barrier. When evaluated on activations drawn from a trained denoising U-Net, the complete system attains a decoder cosine similarity of 0.9906 against an oracle upper bound of 1.0000, while preserving theoretical net energy savings of approximately

10^7\times

over GPU inference. These results constitute the first demonstration of trained-weight, production-scale thermodynamic diffusion inference.

Aditi De

Analog Computing Diffusion Models Thermodynamic Computing Diffusion Model Inference Acceleration

Apr 15, 2026

SocialMirror: Reconstructing 3D Human Interaction Behaviors from Monocular Videos with Semantic and Geometric Guidance

Accurately reconstructing human behavior in close-interaction scenarios is crucial for enabling realistic virtual interactions in augmented reality, precise motion analysis in sports, and natural collaborative behavior in human-robot tasks. Reliable reconstruction in these contexts significantly enhances the realism and effectiveness of AI-driven interactive applications. However, human reconstruction from monocular videos in close-interaction scenarios remains challenging due to severe mutual occlusions, leading local motion ambiguity, disrupted temporal continuity and spatial relationship error. In this paper, we propose SocialMirror, a diffusion-based framework that integrates semantic and geometric cues to effectively address these issues. Specifically, we first leverage high-level interaction descriptions generated by a vision-language model to guide a semantic-guided motion infiller, hallucinating occluded bodies and resolving local pose ambiguities. Next, we propose a sequence-level temporal refiner that enforces smooth, jitter-free motions, while incorporating geometric constraints during sampling to ensure plausible contact and spatial relationships. Evaluations on multiple interaction benchmarks show that SocialMirror achieves state-of-the-art performance in reconstructing interactive human meshes, demonstrating strong generalization across unseen datasets and in-the-wild scenarios. The code will be released upon publication.

Qi Xia, Peishan Cong, Ziyi Wang +6

ShanghaiTech University · Nanyang Technological University · Guangzhou Institute of Energy Conversion, CAS +4

3D Human Reconstruction Monocular 3D Reconstruction Diffusion Models

Apr 15, 2026

Universality of Gaussian-Mixture Reverse Kernels in Conditional Diffusion

We prove that conditional diffusion models whose reverse kernels are finite Gaussian mixtures with ReLU-network logits can approximate suitably regular target distributions arbitrarily well in context-averaged conditional KL divergence, up to an irreducible terminal mismatch that typically vanishes with increasing diffusion horizon. A path-space decomposition reduces the output error to this mismatch plus per-step reverse-kernel errors; assuming each reverse kernel factors through a finite-dimensional feature map, each step becomes a static conditional density approximation problem, solved by composing Norets' Gaussian-mixture theory with quantitative ReLU bounds. Under exact terminal matching the resulting neural reverse-kernel class is dense in conditional KL.

Nafiz Ishtiaque, Syed Arefinul Haque, Kazi Ashraful Alam +1

Center for Mathematics and Interdisciplinary Sciences, Fudan University, Shanghai 200433, China · Shanghai Institute for Mathematics and Interdisciplinary Sciences (SIMIS), Shanghai 200433, China · Northeastern University, Boston, MA 02115, USA +2

Conditional Diffusion Models Diffusion Models Conditional Distribution Estimation

Apr 15, 2026

Diffusion Sequence Models for Generative In-Context Meta-Learning of Robot Dynamics

Accurate modeling of robot dynamics is essential for model-based control, yet remains challenging under distributional shifts and real-time constraints. In this work, we formulate system identification as an in-context meta-learning problem and compare deterministic and generative sequence models for forward dynamics prediction. We take a Transformer-based meta-model, as a strong deterministic baseline, and introduce to this setting two complementary diffusion-based approaches: (i) inpainting diffusion (Diffuser), which learns the joint input-observation distribution, and (ii) conditioned diffusion models (CNN and Transformer), which generate future observations conditioned on control inputs. Through large-scale randomized simulations, we analyze performance across in-distribution and out-of-distribution regimes, as well as computational trade-offs relevant for control. We show that diffusion models significantly improve robustness under distribution shift, with inpainting diffusion achieving the best performance in our experiments. Finally, we demonstrate that warm-started sampling enables diffusion models to operate within real-time constraints, making them viable for control applications. These results highlight generative meta-models as a promising direction for robust system identification in robotics.

Angelo Moroncelli, Matteo Rufolo, Gunes Cagin Aydin +2

University of Applied Science and Arts of Southern Switzerland, Department of Innovative Technologies, IDSIA-SUPSI, Lugano, Switzerland. · Università della Svizzera Italiana, Faculty of Informatics, Lugano, Switzerland. · Politecnico di Milano, Mechanical Department, Milano, Italy.

Robot Dynamics Identification Robot System Identification Robot Dynamics Learning System Identification Diffusion Models ICL for Robotics

Apr 14, 2026

Causal Diffusion Models for Counterfactual Outcome Distributions in Longitudinal Data

Predicting counterfactual outcomes in longitudinal data, where sequential treatment decisions heavily depend on evolving patient states, is critical yet notoriously challenging due to complex time-dependent confounding and inadequate uncertainty quantification in existing methods. We introduce the Causal Diffusion Model (CDM), the first denoising diffusion probabilistic approach explicitly designed to generate full probabilistic distributions of counterfactual outcomes under sequential interventions. CDM employs a novel residual denoising architecture with relational self-attention, capturing intricate temporal dependencies and multimodal outcome trajectories without requiring explicit adjustments (e.g., inverse-probability weighting or adversarial balancing) for confounding. In rigorous evaluation on a pharmacokinetic-pharmacodynamic tumor-growth simulator widely adopted in prior work, CDM consistently outperforms state-of-the-art longitudinal causal inference methods, achieving a 15-30% relative improvement in distributional accuracy (1-Wasserstein distance) while maintaining competitive or superior point-estimate accuracy (RMSE) under high-confounding regimes. By unifying uncertainty quantification and robust counterfactual prediction in complex, sequentially confounded settings, without tailored deconfounding, CDM offers a flexible, high-impact tool for decision support in medicine, policy evaluation, and other longitudinal domains.

Farbod Alinezhad, Jianfei Cao, Gary J. Young +1

Analysis Group · Department of Economics, Northeastern University · Center for Health Policy and Healthcare Research, Northeastern University +1

Counterfactual Prediction Causal Inference Diffusion Models Conditional Diffusion Models

Apr 14, 2026

OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner

The Diffusion Probabilistic Model (DPM) achieves remarkable performance in image generation, while its increasing parameter size and computational overhead hinder its deployment in practical applications. To improve this, the existing literature focuses on obtaining a smaller model with a fixed architecture through model compression. However, in practice, DPMs usually need to be deployed on various devices with different resource constraints, which leads to multiple compression processes, incurring significant overhead for repeated training. To obviate this, we propose a once-for-all (OFA) compression framework for DPMs that yields different subnetworks with various computations in a one-shot training manner. The existing OFA framework typically involves massive subnetworks with different parameter sizes, while such a huge candidate space slows the optimization. Thus, we propose to restrict the candidate subnetworks with a certain set of parameter sizes, where each size corresponds to a specific subnetwork. Specifically, to construct each subnetwork with a given size, we gradually allocate the maintained channels by their importance. Furthermore, we propose a reweighting strategy to balance the optimization process of different subnetworks. Experimental results show that our approach can produce compressed DPMs for various sizes with significantly lower training overhead while achieving satisfactory performance.

Haoyang Jiang, Zekun Wang, Mingyang Yi +6

Renmin University of China Beijing, China · Alibaba Inc. Hangzhou, China · Independent researcher +1

Diffusion Model Compression Model Compression Neural Network Compression Diffusion Models

Apr 14, 2026

SOAR: Self-Correction for Optimal Alignment and Refinement in Diffusion Models

The post-training pipeline for diffusion models currently has two stages: supervised fine-tuning (SFT) on curated data and reinforcement learning (RL) with reward models. A fundamental gap separates them. SFT optimizes the denoiser only on ground-truth states sampled from the forward noising process; once inference deviates from these ideal states, subsequent denoising relies on out-of-distribution generalization rather than learned correction, exhibiting the same exposure bias that afflicts autoregressive models, but accumulated along the denoising trajectory instead of the token sequence. RL can in principle address this mismatch, yet its terminal reward signal is sparse, suffers from credit-assignment difficulty, and risks reward hacking. We propose SOAR (Self-Correction for Optimal Alignment and Refinement), a bias-correction post-training method that fills this gap. Starting from a real sample, SOAR performs a single stop-gradient rollout with the current model, re-noises the resulting off-trajectory state, and supervises the model to steer back toward the original clean target. The method is on-policy, reward-free, and provides dense per-timestep supervision with no credit-assignment problem. On SD3.5-Medium, SOAR improves GenEval from 0.70 to 0.78 and OCR from 0.64 to 0.67 over SFT, while simultaneously raising all model-based preference scores. In controlled reward-specific experiments, SOAR surpasses Flow-GRPO in final metric value on both aesthetic and text-image alignment tasks, despite having no access to a reward model. Since SOAR's base loss subsumes the standard SFT objective, it can directly replace SFT as a stronger first post-training stage after pretraining, while remaining fully compatible with subsequent RL alignment.

You Qin, Linqing Wang, Hao Fei +4

Diffusion Model Alignment Diffusion Model Fine-Tuning Diffusion Models

Apr 14, 2026

DiffusionPrint: Learning Generative Fingerprints for Diffusion-Based Inpainting Localization

Modern diffusion-based inpainting models pose significant challenges for image forgery localization (IFL), as their full regeneration pipelines reconstruct the entire image via a latent decoder, disrupting the camera-level noise patterns that existing forensic methods rely on. We propose DiffusionPrint, a patch-level contrastive learning framework that learns a forensic signal robust to the spectral distortions introduced by latent decoding. It exploits the fact that inpainted regions generated by the same model share a consistent generative fingerprint, using this as a self-supervisory signal. DiffusionPrint trains a convolutional backbone via a MoCo-style objective with cross-category hard negative mining and a generator-aware classification head, producing a forensic feature map that serves as a highly discriminative secondary modality in fusion-based IFL frameworks. Integrated into TruFor, MMFusion, and a lightweight fusion baseline, DiffusionPrint consistently improves localization across multiple generative models, with gains of up to +28% on mask types unseen during fine-tuning and confirmed generalization to unseen generative architectures. Code is available at https://github.com/mever-team/diffusionprint

Paschalis Giakoumoglou, Symeon Papadopoulos

Information Technologies Institute, CERTH

Image Forgery Localization Digital Image Forensics Diffusion Models Image Inpainting

Apr 14, 2026

MAST: Mask-Guided Attention Control for Training-Free Regional-Multi Style Transfer

Style transfer applies the appearance of a reference image to a content image while preserving its spatial structure. Recent diffusion-based methods achieve strong stylization but typically assume a single global style. We instead consider regional-multi style transfer, which assigns multiple references to user-specified regions of a content image. Extending them to this setting reveals two coupled shared-attention issues: ambiguous mass allocation among content and style partitions and degraded selectivity as more styles are jointly normalized, while style aggregation at the attention output further suppresses fine details. We propose MAST (Mask-Guided Attention Control for Training-Free Regional-Multi Style Transfer), a unified attention-control framework for frozen diffusion models. Logit-level Attention Mass Allocation enforces mask-derived partition masses, Sharpness-aware Temperature Scaling adaptively restores selectivity, and Discrepancy-aware Detail Injection recovers high-frequency content. MAST jointly processes all style--mask pairs in a single denoising pass without training, optimization, or post-hoc composition. Across two to five styles, MAST achieves the best average ArtFID, FID, and R-FID among all baselines, demonstrating regional style fidelity, content preservation, and scalability.

Dongkyung Kang, Jaeyeon Hwang, Junseo Park +6

Department of Computer Science & Artificial Intelligence, Dongguk University

Image Style Transfer Attention Control in Diffusion Models Diffusion-Based Image Editing Diffusion Models

Apr 14, 2026

Nucleus-Image: Sparse MoE for Image Generation

We present Nucleus-Image, a text-to-image generation model that establishes a new Pareto frontier in quality-versus-efficiency by matching or exceeding leading models on GenEval, DPG-Bench, and OneIG-Bench while activating only approximately 2B parameters per forward pass. Nucleus-Image employs a sparse mixture-of-experts (MoE) diffusion transformer architecture with Expert-Choice Routing that scales total model capacity to 17B parameters across 64 routed experts per layer. We adopt a streamlined architecture optimized for inference efficiency by excluding text tokens from the transformer backbone entirely and using joint attention that enables text KV sharing across timesteps. To improve routing stability when using timestep modulation, we introduce a decoupled routing design that separates timestep-aware expert assignment from timestep-conditioned expert computation. We construct a large-scale training corpus of 1.5B high-quality training pairs spanning 700M unique images through multi-stage filtering, deduplication, aesthetic tiering, and caption curation. Training follows a progressive resolution curriculum (256 to 512 to 1024) with multi-aspect-ratio bucketing at every stage, coupled with progressive sparsification of the expert capacity factor. We adopt the Muon optimizer and share our parameter grouping recipe tailored for diffusion models with timestep modulation. Nucleus-Image demonstrates that sparse MoE scaling is a highly effective path to high-quality image generation, reaching the performance of models with significantly larger active parameter budgets at a fraction of the inference cost. These results are achieved without post-training optimization of any kind: no reinforcement learning, no direct preference optimization, and no human preference tuning. We release the training recipe, making Nucleus-Image the first fully open-source MoE diffusion model at this quality.

Chandan Akiti, Ajay Modukuri, Murali Nandan Nagarapu +2

T2I Diffusion Models Diffusion Models Diffusion Transformer Diffusion Model Acceleration

Apr 13, 2026

SyncFix: Fixing 3D Reconstructions via Multi-View Synchronization

We present SyncFix, a framework that enforces cross-view consistency during the diffusion-based refinement of reconstructed scenes. SyncFix formulates refinement as a joint latent bridge matching problem, synchronizing distorted and clean representations across multiple views to fix the semantic and geometric inconsistencies. This means SyncFix learns a joint conditional over multiple views to enforce consistency throughout the denoising trajectory. Our training is done only on image pairs, but it generalizes naturally to an arbitrary number of views during inference. Moreover, reconstruction quality improves with additional views, with diminishing returns at higher view counts. Qualitative and quantitative results demonstrate that SyncFix consistently generates high-quality reconstructions and surpasses current state-of-the-art baselines, even in the absence of clean reference images. SyncFix achieves even higher fidelity when sparse references are available.

Deming Li, Abhay Yadav, Cheng Peng +2

Johns Hopkins University · University of Virginia

3D Reconstruction Multi-View 3D Reconstruction Multi-View Consistency Diffusion Models

Apr 13, 2026

Degradation-Aware and Structure-Preserving Diffusion for Real-World Image Super-Resolution

Real-world image super-resolution is particularly challenging for diffusion models because real degradations are complex, heterogeneous, and rarely modeled explicitly. We propose a degradation-aware and structure-preserving diffusion framework for real-world SR. Specifically, we introduce Degradation-aware Token Injection, which encodes lightweight degradation statistics from low-resolution inputs and fuses them with semantic conditioning features, enabling explicit degradation-aware restoration. We further propose Spatially Asymmetric Noise Injection, which modulates diffusion noise with local edge strength to better preserve structural regions during training. Both modules are lightweight add-ons to the adopted diffusion SR framework, requiring only minor modifications to the conditioning pipeline. Experiments on DIV2K and RealSR show that our method delivers competitive no-reference perceptual quality and visually more realistic restoration results than recent baselines, while maintaining a favorable perception--distortion trade-off. Ablations confirm the effectiveness of each module and their complementary gains when combined. The code and model are publicly available at https://github.com/jiyang0315/DASP-SR.git.

Yang Ji, Zonghao Chen, Zhihao Xue +1

KUNBYTE Hangzhou, Zhejiang Province, China · GOLDMYE Xuhui District, Shanghai, China

Degradation-Aware Image Restoration Image Super-Resolution Diffusion Models for Image Restoration Image Restoration Diffusion Models

Apr 13, 2026

Learning Discrete Diffusion of Graphs via Free-Energy Gradient Flows

Diffusion-based models on continuous spaces have seen substantial recent progress through the mathematical framework of gradient flows, leveraging the Wasserstein-2 (

{W}_2

) metric via the Jordan-Kinderlehrer-Otto (JKO) scheme. Despite the increasing popularity of diffusion models on discrete spaces using continuous-time Markov chains, a parallel theoretical framework based on gradient flows has remained elusive due to intrinsic challenges in translating the

{W}_2

distance directly into these settings. In this work, we propose the first computational approach addressing these challenges, leveraging an appropriate metric

W_K

on the simplex of probability distributions, which enables us to interpret widely used discrete diffusion paths, such as the discrete heat equation, as gradient flows of specific free-energy functionals. Through this theoretical insight, we introduce a novel methodology for learning diffusion dynamics over discrete spaces, which recovers the underlying functional directly by leveraging first-order optimality conditions for the JKO scheme. The resulting method optimizes a simple quadratic loss, trains extremely fast, does not require individual sample trajectories, and only needs a numerical preprocessing computing

W_K

-geodesics. We validate our method through extensive numerical experiments on synthetic data, showing that we can recover the underlying functional for a variety of graph classes.

Dario Rancati, Jan Maas, Francesco Locatello

Institute of Science and Technology Austria

Diffusion Models Graph Diffusion Wasserstein Gradient Flows

Apr 13, 2026

Energy-oriented Diffusion Bridge for Image Restoration with Foundational Diffusion Models

Diffusion bridge models have shown great promise in image restoration by explicitly connecting clean and degraded image distributions. However, they often rely on complex and high-cost trajectories, which limit both sampling efficiency and final restoration quality. To address this, we propose an Energy-oriented diffusion Bridge (E-Bridge) framework to approximate a set of low-cost manifold geodesic trajectories to boost the performance of the proposed method. We achieve this by designing a novel bridge process that evolves over a shorter time horizon and makes the reverse process start from an entropy-regularized point that mixes the degraded image and Gaussian noise, which theoretically reduces the required trajectory energy. To solve this process efficiently, we draw inspiration from consistency models to learn a single-step mapping function, optimized via a continuous-time consistency objective tailored for our trajectory, so as to analytically map any state on the trajectory to the target image. Notably, the trajectory length in our framework becomes a tunable task-adaptive knob, allowing the model to adaptively balance information preservation against generative power for tasks of varying degradation, such as denoising versus super-resolution. Extensive experiments demonstrate that our E-Bridge achieves state-of-the-art performance across various image restoration tasks while enabling high-quality recovery with a single or fewer sampling steps. Our project page is https://jinnh.github.io/E-Bridge/.

Jinhui Hou, Zhiyu Zhu, Junhui Hou

Department of Computer Science, City University of Hong Kong · Department of Computer Science, City University of Hong Kong (Dongguan)

Diffusion Models for Image Restoration Diffusion Models Diffusion Model Inference Acceleration Image Super-Resolution

Apr 13, 2026

Tracking High-order Evolutions via Cascading Low-rank Fitting

Diffusion models have become the de facto standard for modern visual generation, including well-established frameworks such as latent diffusion and flow matching. Recently, modeling high-order dynamics has emerged as a promising frontier in generative modeling. Rather than only learning the first-order velocity field that transports random noise to a target data distribution, these approaches simultaneously learn higher-order derivatives, such as acceleration and jerk, yielding a diverse family of higher-order diffusion variants. To represent higher-order derivatives, naive approaches instantiate separate neural networks for each order, which scales the parameter space linearly with the derivative order. To overcome this computational bottleneck, we introduce cascading low-rank fitting, an ordinary differential equation inspired method that approximates successive derivatives by applying a shared base function augmented with sequentially accumulated low-rank components. Theoretically, we analyze the rank dynamics of these successive matrix differences. We prove that if the initial difference is linearly decomposable, the generic ranks of high-order derivatives are guaranteed to be monotonically non-increasing. Conversely, we demonstrate that without this structural assumption, the General Leibniz Rule allows ranks to strictly increase. Furthermore, we establish that under specific conditions, the sequence of derivative ranks can be designed to form any arbitrary permutation. Finally, we present a straightforward algorithm to efficiently compute the proposed cascading low-rank fitting.

Zhao Song

Simons Institute for Theory of Computing, UC Berkeley

Diffusion Models

Apr 12, 2026

Query Lower Bounds for Diffusion Sampling

Diffusion models generate samples by iteratively querying learned score estimates. A rapidly growing literature focuses on accelerating sampling by minimizing the number of score evaluations, yet the information-theoretic limits of such acceleration remain unclear. In this work, we establish the first score query lower bounds for diffusion sampling. We prove that for

d

-dimensional distributions, given access to score estimates with polynomial accuracy

\varepsilon=d^{-O(1)}

(in any

L^p

sense), any sampling algorithm requires

\widetildeΩ(\sqrt{d})

adaptive score queries. In particular, our proof shows that, within any polynomial total-query budget, successful sampling requires searching over

\widetildeΩ(\sqrt{d})

distinct noise levels, providing a formal explanation for why multiscale noise schedules are necessary in practice.

Zhiyang Xun, Eric Price

UT Austin

Query Complexity Diffusion Sampling Information-Theoretic Lower Bounds Diffusion Models