Generative Adversarial Networks and Image Synthesis

Derniers papiers

• Consistent Distributed Ranking of Generative Models via Kernel Distances

  Zixiao Wang, Farzan Farnia, Zhenghao Lin, Yunheng Shen, Bei Yu · 26 juin 2026

  Ranking generative models based on the fidelity and diversity of their outputs is required to identify the best generator in a group of candidate generative AI models. To rank a group of models in a conventional centralized setting, a standard score is commonly evaluated for each involved model. The…

• Anatomy-Guided Residual Motion Diffusion for Controllable 4D Cardiac MRI Synthesis

  Yiheng Cao, Gustavo Andrade-Miranda, Jiatian Zhang, Lingxiao Zhao, Xin Gao · 26 juin 2026

  Developing robust artificial intelligence models for 4D (3D + time) medical imaging is constrained by limited annotated data, inter-device domain shifts, and privacy restrictions. To address this, we propose a 4D controllable generative framework for anatomically consistent data augmentation. A semi…

• Disco-LoRA: Disentangled Composition of Content, Style, and Motion for Multi-concept Video Customization

  Xuancheng Xu, Gengyun Jia, Bing-Kun Bao · 26 juin 2026

  Video customization based on Text-to-Video (T2V) models aims to learn specific features from reference data to generate controllable videos. While significant strides have been made in image stylization and video motion customization, simultaneously controlling multiple concepts, such as content, st…

• No Free Lunch: Non-Asymptotic Analysis of Prediction-Powered Inference

  Pranav Mani, Peng Xu, Zachary C. Lipton, Michael Oberst · 26 juin 2026

  Prediction-Powered Inference (PPI) is a popular strategy for combining gold-standard and possibly noisy pseudo-labels to perform statistical estimation. Prior work has shown an asymptotic \enquote{free lunch} for PPI++, an adaptive form of PPI, showing that the \textit{asymptotic} variance of PPI++ …

• Beyond the Hard Budget: Sparsity Regularizers for More Interpretable Top-k Sparse Autoencoders

  Nathana\"el Jacquier, Maria Vakalopoulou, Mahdi S. Hosseini · 26 juin 2026

  Sparse autoencoders (SAEs) have become a leading tool for interpreting the representations of vision foundation models, decomposing their polysemantic activations into a larger set of sparse, more monosemantic features. The Top-$k$ SAE, a now-standard variant, enforces sparsity architecturally throu…

• CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention

  Sayak Dutta · 26 juin 2026

  Recurrent models must forget in order to remember, yet the state of the art decides what to erase without consulting what is stored -- the gate sees only the arriving token, not the memory it is about to modify. This memory-blind gating is one of three coupled defects in the leading delta-rule archi…

• Hallucination in World Models is Predictable and Preventable

  Nicklas Hansen, Xiaolong Wang · 26 juin 2026

  Modern generative world models render increasingly realistic action-controllable futures, yet they frequently hallucinate: rollouts remain visually fluent while drifting from the ground-truth dynamics. We hypothesize that hallucination concentrates in low-coverage regions of the state-action space, …

• Dataset Usage Inference without Shadow Models or Held-out Data

  Wojciech {\L}apacz, Stanis{\l}aw Pawlak, Jan Dubi\'nski, Franziska Boenisch, Adam Dziedzic · 26 juin 2026

  How much of my data was used to train a machine learning model? Dataset Usage Inference (DUI) aims to answer this by estimating what fraction of a dataset contributed to a model's training. However, existing DUI methods rely on assumptions that rarely hold in practice: they require training expensiv…

• Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis

  Abrar Alotaibi, Moataz Ahmed · 26 juin 2026

  Neural Architecture Search (NAS) has emerged as a pivotal technique in optimizing the design of Generative Adversarial Networks (GANs), automating the search for effective architectures while addressing the challenges inherent in manual design. This paper provides a comprehensive review of NAS metho…

• Scaling Multi-Reference Image Generation with Dynamic Reward Optimization

  Wenwang Huang, Yusen Fu, Junjie Wang, Mengfei Huang, Yulin Li, Gan Liu, Jing Cai, Yancheng He, Zhuotao Tian · 26 juin 2026

  While personalized image generation has achieved remarkable progress, multi-reference image generation (MRIG) remains a challenging task. Most existing benchmarks fail to adequately evaluate complex MRIG scenarios, hindering further progress in this area. To better assess model performance on comple…

• LearniBridge: Learnable Calibration of Feature Caching for Diffusion Models Acceleration

  Xuyue Huang, Zhe Chen, Wang Shen, Xiao-Ping Zhang · 26 juin 2026

  Diffusion Transformers (DiTs) have driven substantial progress in image and video generation but suffer from prohibitive computational costs. Feature caching accelerates inference by reusing intermediate representations. Existing methods rely on historical features for implementation simplicity, yet…

• DanceOPD: On-Policy Generative Field Distillation

  Wei Zhou, Xiongwei Zhu, Zelin Xu, Bo Dong, Lixue Gong, Yongyuan Liang, Meng Chu, Leigang Qu, Lingdong Kong, Wei Liu, Tat-Seng Chua · 26 juin 2026

  Modern image generation demands a single model that unifies diverse capabilities, including text-to-image (T2I), local editing, and global editing. However, these capabilities are rarely naturally aligned and often conflict. For instance, editing tends to degrade T2I performance, while global and lo…

• Perception, Verdict, and Evolution: Hindsight-Driven Self-Refining Forensics Agent for AI-Generated Image Detection

  Yangjun Wu, Keyu Yan, Yu Liu, Jingren Zhou, Fei Huang, Rong Zhang, Zhou Zhao, Fei Wu · 26 juin 2026

  The rapid advancement of generative models presents a significant challenge to existing deepfake detection methods, particularly given the widespread dissemination of highly realistic AI-generated images. Although Multimodal Large Language Models (MLLMs) show strong potential for this task, existing…

• LCG: Long-Context Consistent Image Generation with Sparse Relational Attention

  Zihao Wang, Yijia Xu, Haoze Zheng, Xuran Ma, Haokun Gui, Harry Yang · 26 juin 2026

  Recent image generation models achieve impressive quality in single-image synthesis, but often fail to maintain consistency across sequential outputs, as required in comics, storyboards, and visual narratives. We propose Long-Context Generation (LCG), a framework for long-context multi-image text-to…

• Safe Autoregressive Image Generation with Iterative Self-Improving Codebooks

  Yunqi Xue, Zhijiang Li, Philip Torr, Jindong Gu · 26 juin 2026

  Unlike diffusion-based models that operate in continuous latent spaces, autoregressive unified multimodal models produce images by sequentially predicting discretized visual tokens. These tokens are derived from a codebook that maps embeddings to quantized visual patterns. The language-like architec…

• Concept Removal for Frontier Image Generative Models

  Aditya Kumar, Pierre Joly, Adam Dziedzic, Franziska Boenisch · 25 juin 2026

  Image generative models are trained on massive, largely uncurated internet-scale datasets that contain undesirable visual concepts. Efficiently removing such concepts from the model generations without degrading the quality of output images remains challenging. We introduce a novel concept removal m…

• Sample complexity of unbalanced entropic OT

  Francisco Andrade, Gabriel Peyr\'e, Clarice Poon · 25 juin 2026

  Optimal transport (OT) has become a central language for comparing probability measures, but exact balanced OT is often both too rigid for data with missing, created, or destroyed mass and subject to unfavorable high-dimensional sample complexity. Entropic regularization and unbalanced relaxations a…

• EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data Synthesis

  Huaqiu Li, Jiahao Wang, Sijia Cai, Hualian Sheng, Bing Deng, Jieping Ye, Wenhan Luo · 25 juin 2026

  While image stylization has been studied extensively, video stylization remains a critical and largely unsolved challenge in the field of intelligent content creation. Existing methods, usually utilizing a reference image as the style prior, suffer from content leakage, data scarcity and limited ada…

• Variational Autoencoder Layer

  Gananath R · 25 juin 2026

  Variational Autoencoders (VAEs) belong to a family of autoencoders with probabilistic properties, making them well suited for generating data by producing a smooth and continuous latent space. Despite being introduced over a decade ago, the method continues to be widely adopted in both research and …

• Causal-rCM: A Unified Teacher-Forcing and Self-Forcing Open Recipe for Autoregressive Diffusion Distillation in Streaming Video Generation and Interactive World Models

  Kaiwen Zheng, Guande He, Min Zhao, Jintao Zhang, Huayu Chen, Jianfei Chen, Chen-Hsuan Lin, Ming-Yu Liu, Jun Zhu, Qianli Ma · 25 juin 2026

  Autoregressive video diffusion with causal diffusion transformers has emerged as a major paradigm for real-time streaming video generation and action-conditioned interactive world models. In this work, we extend rCM, an advanced diffusion distillation framework, to autoregressive video diffusion. Th…

• Why Do Accumulated Transformations Extrapolate?

  Mahesh Godavarti · 25 juin 2026

  PaTH Attention showed that replacing RoPE's position-indexed rotations with accumulated data-dependent Householder reflections yields strong length extrapolation, though performance degrades at extreme context lengths. We ask whether this depends on Householder-specific structure or reflects a gener…

• Wan-Streamer v0.1: End-to-end Real-time Interactive Foundation Models

  Lianghua Huang, Zhifan Wu, Wei Wang, Yupeng Shi, Mengyang Feng, Junjie He, Chenwei Xie, Yu Liu, Jingren Zhou, Ang Wang, Bang Zhang, Baole Ai, Chen Liang, Cheng Yu, Chongyang Zhong, Jinwei Qi, Kai Zhu, Pandeng Li, Peng Zhang, Wenyuan Zhang, Xinhua Cheng, Yitong Huang, Yun Zheng, Zoubin Bi · 25 juin 2026

  We present Wan-Streamer, a native-streaming, end-to-end interactive foundation model designed from the ground up for real-time, low-latency, full-duplex audio-visual interaction. Wan-Streamer seamlessly models language, audio, and video as both input and output within a single Transformer, where the…

• Efficient Adaptive Data Acquisition via Pretrained Belief Representations

  Daolang Huang, Zhuoyue Huang, Conor Hassan, Luigi Acerbi, Samuel Kaski, Tom Rainforth · 25 juin 2026

  Learning effective policies for adaptive data acquisition remains challenging: posterior-based methods rely on surrogate models and posterior approximations that can be misspecified or biased, while direct policy-learning methods map from historical observations and fail to exploit available model r…

• Latent Block-Diffusion Temporal Point Processes: A Semi-Autoregressive Framework for Asynchronous Event Sequence Generation

  Shuai Zhang, Yancheng Chen, Chuan Zhou, Yang Liu, Xixun Lin, Xiangyu Zhao, Jun Zhu, Zhi-Ming Ma · 25 juin 2026

  Modeling and sampling from the underlying distribution of asynchronous event sequences are crucial in various real-world applications, including social networks, medical diagnosis, and financial transactions. Existing autoregressive methods suffer from error accumulation during multi-step generation…

• A Time-Reparameterized Cumulative Intensity Extrapolation Sampler for Discrete Flow Matching

  Feiyang Fu, Hehe Fan · 24 juin 2026

  Discrete flow matching (DFM) provides a principled framework for generative modeling on discrete state spaces via continuous-time Markov chain dynamics. In practice, sampling for DFM commonly employs discretizations such as $\tau$-leaping, yet efficient sampling methods under a limited number of fun…