Generative Adversarial Networks and Image Synthesis

Derniers papiers

• Disagree to Accelerate: Closing the Loop on Diffusion Feature Forecasts

  Yanchao Li, Jiaqing Xie, Ben Gao, Wanhao Liu, Yanbo Wang, T. Y. Tsui, Jinfei Liu, Yuqiang Li, Tianfan Fu · 4 août 2026

  Training-free feature forecasting accelerates diffusion sampling by predicting features at skipped denoising steps. Recent work has mainly focused on designing stronger forecasters. Yet forecast error varies sharply across steps, and open-loop caches trust the forecast in full at every skipped step.…

• WorldDynCache: Risk-Controlled Latent Dynamics Approximation for Diffusion World Model

  Leyang Chen, Junyi Wu, Shaoqiu Zhang, Yulun Zhang · 4 août 2026

  Diffusion world models generate high-quality futures, but re- peated transformer evaluations make inference prohibitively slow. Existing caches reuse intermediate features, selectively update tokens, or reuse and extrapolate denoising outputs ac- cording to local drift or short native-space historie…

• Exploring More to Solve More: Boosting Diversity in Text Diffusion Models via Entropy-Based Guidance

  Jingwei Zhang, Haoyu Lei, Zijin Feng, Jiacheng Sun, Farzan Farnia · 4 août 2026

  Although diffusion models have revolutionized continuous domains like image synthesis through high quality generations and controllable guidance mechanisms, bringing this controllability to the discrete, sequential nature of text remains an open challenge. Meanwhile, current sampling strategies and …

• One-Sided Quantile Coupling for Flow Matching

  Jin-Young Kim, So-Yoon Cho, Hyun-Gyoon Kim · 4 août 2026

  Flow Matching trains continuous-time generative models by regressing the velocity field of a probability path between a simple source distribution and a target data distribution. The coupling that pairs source and target samples strongly affects optimization and sample quality, but structured coupli…

• Face-D(^2)CL: Multi-Domain Synergistic Representation with Dual Continual Learning for Facial DeepFake Detection

  Yushuo Zhang, Yu Cheng, Yongkang Hu, Jiuan Zhou, Jiawei Chen, Zhaoxia Yin · 4 août 2026

  Facial forgery techniques are advancing rapidly, posing severe threats to public trust and information security while imposing higher demands on the continual adaptation of DeepFake detection models. Although continual learning enables models to adapt to emerging forgery methods, existing approaches…

• xPress: Parallel Refinement for Diffusion Drafters in Speculative Decoding

  Zheng Wang, Davis Wertheimer, Yu Chin Fabian Lim, Mudhakar Srivatsa, Raghu K. Ganti, Minjia Zhang, Naigang Wang · 4 août 2026

  Block-diffusion drafters like dFlash generate an entire block of draft tokens in a single forward pass, drastically reducing the overhead of multiple-token drafting in speculative decoding. The crucial final step of the single-pass discrete denoising process involves using the logit distribution at …

• Round-Trip Consistency: Bidirectional Diffusion Models Can Predict Their Own Rollout Errors

  Alexander Scheinker · 4 août 2026

  Autoregressive models accumulate error over long rollouts, yet at deployment there is no ground truth to measure it against. We train a single conditional latent diffusion model that steps a dynamical system forward or backward in time via a direction flag, and show that this bidirectionality suppli…

• UDT: Reconciling U-Nets and Diffusion Transformers with Data-Adaptive Token Reduction

  Junno Yun, Ya\c{s}ar Utku Al\c{c}alar, Mehmet Ak\c{c}akaya · 4 août 2026

  Diffusion Transformers (DiTs) have emerged as a core architecture in generative modeling due to their scalability and adaptability to multimodal tasks. DiTs comprise isotropic transformer blocks, and learn representations progressively across depth, where the denoising objective drives later layers …

• DAVET: Denoising-Aware Visual Evidence Trajectory Allocation for Diffusion Vision-Language Models

  Yongkang Zhou, Xiang Xia, Cheng Yan, Fan Xu, Wuyang Zhang · 4 août 2026

  Diffusion vision-language models (dVLMs) iteratively denoise masked responses while conditioning each denoising step on visual evidence, making visual conditioning a substantial recurring inference cost. Unlike autoregressive decoding, diffusion generation repeatedly revisits the entire response as …

• Beckmann Transport Models: From Autonomous Flows to One-Step Maps

  Lee Cheuk-Kit, Florentin Coeurdoux, Peter Potaptchik, Yilun Du, Michael Samuel Albergo, Eric Vanden-Eijnden · 4 août 2026

  We propose an instantiation of flow matching that relies on a time-independent velocity field (an \emph{autonomous flow}) to exactly map between two distributions, so long as the target is singular, i.e.\ supported on a lower-dimensional data manifold. We also show that the one-step generative map a…

• DiffusionGemma Technical Report

  DiffusionGemma Team, Adrien Ali Ta\"iga, James Assiene, Daniele Calandriello, Rahma Chaabouni, Jo\~ao Gante, Tamara von Glehn, Nate Keating, Chris Knutsen, Martin Kukla, Tianlin Liu, Ivan Lobov, Ofir Nabati, Jo\~ao Gabriel Oliveira, Nicolas Perez-Nieves, Nastasia Prutianova, Bobak Shahriari, Jean Tarbouriech, Pavel Tyletski, \c{C}a\u{g}lar \"Unl\"u, Cindy Wu, Glenn Cameron, Jerome Connor, Sertan Girgin, Maarten Grootendorst, Alon Levkovitch, Eliya Nachmani, Omar Sanseviero, Piotr Stanczyk, Quentin Berthet, Andrew Campbell, Cl\'ement Crepy, Valentin De Bortoli, Arnaud Doucet, Romuald Elie, Alexandre Galashov, Klaus Greff, Alexis Jacq, David Ruhe, Yu-Han Wu, Sebastian Flennerhag, Brendan O'Donoghue, George Scrivener, Shantanu Thakoor · 4 août 2026

  We introduce DiffusionGemma, an experimental open-weight language model that uses discrete diffusion to generate text at exceptionally high speed. Rather than decoding one token at a time, DiffusionGemma iteratively refines blocks of 256 tokens in parallel, avoiding the sequential decoding bottlenec…

• SPARE: Structural Parameter-Free Affinity Regularization for Flow Matching

  Zong-Wei Hong, Jinglun Li, Shen Zhang, Yuhan Liu, Linze Li, Yao Tang · 4 août 2026

  Denoising diffusion transformers achieve strong generation quality but converge slowly during training. Regularizing their internal representations has emerged as an effective accelerator, yet existing methods split into two families with complementary costs. Target-based methods strengthen represen…

• WaiT for the Signal: Simple Frequency-Aware Flow-Matching

  Krunoslav Lehman Pavasovic, Th\'eophane Vallaeys, St\'ephane Mallat, Giulio Biroli, Luke Zettlemoyer, Brian Karrer, Jakob Verbeek · 3 août 2026

  As image generation models scale to ever higher resolutions, global coherence, local detail, and texture fidelity become critical axes for generation quality. However, standard flow matching treats all spatial frequencies uniformly, ignoring the natural frequency hierarchy where high-frequency bands…

• MBDiff: Multi-view Behavior-aware Diffusion Model for Probabilistic Utility Data Imputation

  Rongchao Xu, Lin Jiang, Dahai Yu, Ximiao Li, Guang Wang · 3 août 2026

  Utility data (e.g., electricity, water, and gas consumption), collected by ubiquitous sensors and embedded devices, often contains substantial missing values due to various factors such as device failures and data transmission issues. The data missingness can severely impact utility billing accuracy…

• Detecting AI-Generated Videos with Spiking Neural Networks

  Minsuk Jang, Yujin Yang, Hee-Seon Kim, Minseok Son, Younghun Kim, Changick Kim · 3 août 2026

  Modern AI-generated videos are photorealistic at the single-frame level, leaving inter-frame dynamics as the main remaining axis for detection. Existing detectors typically handle this temporal evidence in three ways: feeding the full frame sequence to a generic temporal backbone, reducing one domin…

• Conditioning Tree-Based Diffusions and Flows for Probabilistic Tabular Regression

  Silas Koemen · 3 août 2026

  Tree-based diffusion models fit flexible conditional predictive distributions for tabular regression without a neural density estimator, but they inherit their design defaults---noising path, parameterization, training distribution, features, sampler---from the neural setting. We show these defaults…

• FairDiffuseVQVAE: Sampling-Time Fairness in Tabular Diffusion via Conditional Refinement of Vector-Quantized Latents

  Nitish Nagesh, Mahdi Bagheri, Amir M. Rahmani · 3 août 2026

  Synthetic tabular data is increasingly used in privacy-preserving data sharing, data augmentation, and to mitigate downstream classifier bias. State-of-the-art tabular diffusion models such as TabDDPM and TabSyn achieve excellent distributional fidelity but offer no mechanism for fairness; conversel…

• Mitigating Compounding Error via Video Representation Regularization

  Taiye Chen, Qi Zhang, Yisen Wang · 30 juillet 2026

  Video diffusion-based world models enable long autoregressive video generation for robotics, autonomous driving and simulation tasks, yet sliding-window autoregressive inference suffers from severe error accumulation that degrades frame quality over time. Although this phenomenon has been widely obs…

• Amortized Moment Matching for Visual Generation

  Wenze Liu, Xintao Wang, Pengfei Wan, Xiangyu Yue · 30 juillet 2026

  We propose amortized moment matching, utilizing neural networks to learn data moments as distributional training signals. By casting diffusion denoisers through polynomial projections, we establish a general framework for moment amortization, revealing that an $n$-th degree projection explicitly ide…

• Dual Inversion for Text-to-Image Diffusion Models: From Both Prompt and Noise Perspectives

  Xiaolong Liu, Junjian Li, Yuan Xiao, Jiaqi Deng, Dayong Ye, Tianqing Zhu, Huan Huo · 30 juillet 2026

  Prompt inversion, as a typical reverse engineering technique, enables text-to-image (T2I) diffusion models to generate the desired target images without extensive prompt engineering. However, existing prompt inversion methods suffer from significant limitations: (1) gradient-based methods are unstab…

• FloDR: An invertible dimensionality reduction method based on a normalising flow

  Abdallah Baraka, Daniel Probst · 30 juillet 2026

  It is common for two-dimensional embeddings of high-dimensional data to be read far beyond what they can support. Distances in and between clusters, the meaning behind empty spaces, and the amount of structure hidden at each point are generally invisible in the output of methods such as t-SNE and UM…

• Flow Map Learning via Nongradient Vector Flow

  Mark Goldstein, Anshuk Uppal, Raghav Singhal, Aahlad Puli, Rajesh Ranganath · 30 juillet 2026

  Diffusion and flow-based models benefit from simple regression losses, but inference incurs significant overhead because sampling requires integration. Consistency models address this by directly learning the flow maps along the ODE trajectory, opening a design space between one-step and many-step a…

• From Interface to Inference: Eliciting Any-Order Inference from Any-Order Models

  Seunggeun Kim, Jaeyeon Kim, Taekyun Lee, Yuyuan Chen, Yilun Du, Sham Kakade, Sitan Chen · 30 juillet 2026

  Many discrete reasoning tasks, such as code generation, are inherently non-causal: programmers move between high-level structure and local details, a process we call any-order inference. For autoregressive language models, which lack a native any-order interface, non-causal abilities such as infilli…

• PreDiff-LM: Pretrained Discrete Masked Diffusion Language Modeling with Hybrid Attention

  Zhengtao Yao, Runhao Li, Xupeng Chen, Jiayi Cheng, Chenqian Le, Michael Yue, Jesson Wang, Siheng Wang, Guang Yang, Haoyan Xu, Chenhao Wei, Zhengqing Yuan, Youran Shen, Yanfang Ye, Junhao Dong · 29 juillet 2026

  Discrete masked diffusion language models support bidirectional generation and infilling, but adapting pretrained autoregressive (AR) transformers requires reconciling causal pretraining with bidirectional denoising. We study this problem at the level of attention rather than claiming AR-weight reus…

• Parallel Decoding Distillation for Fast Image and Video Generation

  Neta Shaul, Chao Liu, Arash Vahdat, Julius Berner · 29 juillet 2026

  Generation in video diffusion or flow models is computationally expensive due to the slow and iterative sampling process. Current state-of-the-art (SOTA) acceleration methods heavily rely on variational score distillation (VSD) and adversarial losses to distill diffusion models into few-step generat…