Generative Adversarial Networks and Image Synthesis

Derniers papiers

• Combating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning

  Xuehui Wang, Xuankun Yang, Wei Shen · 3 juillet 2026

  Visual token pruning is a crucial strategy for accelerating VLMs by compressing redundant image patches, yet existing methods often fail to preserve critical cues under dense instructions and fine-grained queries. In this paper, we investigate this failure and identify two underlying bottlenecks: th…

• Fourier Preconditioning for Neural Feature Learning

  Preston Pitzer, Anish Pradhan, Harpreet S. Dhillon · 3 juillet 2026

  Mutual information (MI)-inspired feature learning techniques are capable of generating low-dimensional embeddings that retain nonlinear dependence structures, but direct estimations of MI suffer from noisy probability distribution estimates in the low-data regime. The H-Score objective, computed fro…

• Optimizing Visual Generative Models via Distribution-wise Rewards

  Ruihang Li, Mengde Xu, Shuyang Gu, Leigang Qu, Fuli Feng, Han Hu, Wenjie Wang · 3 juillet 2026

  Conventional reinforcement learning strategies for visual generation typically employ sample-wise reward functions, yet this practice frequently results in reward hacking that degrades image diversity and introduces visual anomalies. To address these limitations, we present a novel framework that fi…

• NeoMap: Training-free Novel-View Synthesis from Single Images and Videos

  Jinxi Li, Tianyi Zhang, Yafei Yang, Zihui Zhang, Peng Huang, Koon Wing Macgyver Lin, Bo Yang · 3 juillet 2026

  We study the challenging problem of novel view video synthesis from single images or monocular videos. Existing methods, which operate under the assumption that pre-trained video models lack native novel view synthesis capability and enforce view alignment via camera conditioning, task-specific fine…

• Expander Sparse Autoencoders: Parameter-Efficient Dictionaries for Mechanistic Interpretability

  Rodrigo Mendoza-Smith · 3 juillet 2026

  Sparse autoencoders (SAEs) decompose internal activations of neural networks into sparse linear combinations of learned features by fitting an overcomplete dictionary $\mathbf{W}\in\mathbb{R}^{m\times n}$ with $m<n$, and inferring a sparse code $\mathbf{x}\in\mathbb{R}^n$ from $\mathbf{h}\approx\mat…

• eXact-Prior Variational Autoencoder (X-VAE): Learning Data-Adaptive Gaussian Mixture Priors for Latent Distributions

  Qijun Chen, Shaofan Li · 3 juillet 2026

  Variational Autoencoders (VAEs) commonly assume a standard isotropic Gaussian prior over the latent space, an assumption that often fails to capture the true distribution of latent representations for complex datasets. This mismatch can limit reconstruction accuracy, reduce sample quality, and const…

• Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Models

  Enrico Cassano, Riccardo Renzulli, Rayyan Ahmed, Marco Grangetto, Stephan Alaniz · 3 juillet 2026

  Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated features can serve as controllable intervention points. In this work, we systematically evaluate this assumption in the context of object erasure and steer…

• Set Diffusion: Interpolating Token Orderings Between Autoregression and Diffusion for Fast and Flexible Decoding

  Marianne Arriola, Volodymyr Kuleshov · 3 juillet 2026

  Discrete diffusion models have steadily improved in quality relative to autoregressive (AR) models. However, these models are normally constrained to fixed-length generation and do not support key-value (KV) caching. Block diffusion partially bridges diffusion and AR by generating token blocks left-…

• Population-Based Multi-Objective Training of Discriminators for Semi-Supervised GANs

  Francisco Sede\~no, Francisco Chicano, Jamal Toutouh · 3 juillet 2026

  Semi-supervised generative adversarial networks (SSL-GANs) can exploit large unlabeled datasets while retaining a classifier in the discriminator, but their training is often unstable. This paper proposes a population-based evolutionary training strategy in which discriminator learning is formulated…

• SUNTA: Hierarchical Video Prediction with Surprise-based Chunking

  Tomoshi Iiyama, Masahiro Suzuki, Yutaka Matsuo · 3 juillet 2026

  Hierarchical state-space models (HSSMs) offer a promising approach to long-horizon prediction by segmenting sequences into temporal chunks. However, their performance hinges on how chunk boundaries are determined. While prior HSSMs typically rely on fixed-length chunking or similarity-based boundary…

• Locality-Aware Continual Unlearning for Diffusion Models

  Naveen George, Naoki Murata, Yuhta Takida, Konda Reddy Mopuri, Yuki Mitsufuji · 3 juillet 2026

  Real-world deployment of text-to-image diffusion models requires continual concept removal as new privacy, copyright, or safety obligations arise over time. Existing unlearning methods, however, are designed for single-step deletion and collapse after only 3-5 sequential applications. We trace this …

• Model Merging as Probabilistic Inference in Fine-Tuning Parameter Space

  Long Minh Bui, Tuan Anh Le Van, Tung Phi Duc, Phi Le Nguyen, Jana Doppa, Trong Nghia Hoang · 3 juillet 2026

  Model merging aims to combine existing single-task solutions into a multi-task solution without additional data-driven fine-tuning.~Most existing approaches achieve this using geometric properties of local solution spaces. However, such geometric views provide limited guidance for scoring how statis…

• AbsoluteDegradation: A Physics-Inspired Synthetic Film-Degradation Pipeline and Archival Film Restoration Benchmark

  Miko{\l}aj Jastrz\k{e}bski, Dawid Glinkowski, Dawid Zieli\'nski, Daniel Borkowski, Wojciech Koz{\l}owski, Kamil Adamczewski · 3 juillet 2026

  Restoring archival film remains a fundamentally challenging problem due to the absence of paired training data and the lack of standardized evaluation benchmarks. Pristine versions of deteriorated footage are physically unrecoverable, requiring supervised methods to rely on synthetic data that often…

• SNAP-FM: Sparse Nonlinear Accelerated Projection for Physics-Constrained Generative Modeling

  Alaina Kolli, Theodoros Xenakis, Utkarsh Utkarsh, Pengfei Cai, Rafael Gomez-Bombarelli, Alan Edelman, Christopher Vincent Rackauckas · 2 juillet 2026

  Generative models have emerged as scalable surrogates for physical simulation, yet they offer no guarantee that their outputs respect the conservation laws, boundary conditions, and nonlinear invariants that govern the underlying physics. Constrained sampling closes this gap, enforcing such constrai…

• UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios

  Tian Ye, Song Fei, Lei Zhu · 2 juillet 2026

  Diffusion transformers have recently delivered strong text-to-image generation around 1K resolution, but we show that extending them to native 4K across diverse aspect ratios exposes a tightly coupled failure mode spanning positional encoding, VAE compression, and optimization. Tackling any of these…

• MEPA: Multi-Scale Representation Alignment for Visual Autoregressive Modeling with Mixture of Experts

  Nuoyan Zhou, Zhijun Tu, Lei Yu, Kun Cheng, Jie Hu, Nannan Wang, Xinghao Chen · 2 juillet 2026

  Visual AutoRegressive modeling (VAR) has pioneered a coarse-to-fine multi-scale autoregressive generative paradigm, demonstrating strong capabilities in image generation. However, VAR still suffers from inherent deficiencies in multi-scale representation learning. Specifically, lower scales primaril…

• RetailSMV: Exocentric vs. Egocentric Adaptation of Foundation Video World Models in Retail

  Amirreza Rouhi, Rajat Aggarwal, Parikshit Sakurikar, Anoop M. Namboodiri, Sashi P. Reddi · 2 juillet 2026

  Foundation video diffusion models are increasingly viewed as world simulators for embodied agents, yet their pretraining on internet-scale generic video leaves them poorly aligned with real-world deployment domains. We study parameter-efficient adaptation of a pretrained foundation video world model…

• Sequentially-Controlled Interactive Multi-Particle Flow-Maps for Online Feedback-Driven Search

  Binglin Ji, Anindya Sarkar, Hengchang Lu, Jens Sj\"olund, Yevgeniy Vorobeychik · 2 juillet 2026

  While generative models have enabled training-free reward alignment, current methods typically excel in local exploration within narrow regions of the underlying distribution. These approaches struggle when preferences are unknown a priori and only revealed through sequential feedback-a scenario dem…

• Accelerating Discrete Diffusion Models with Parallel-In-Time Sampling

  Yu Yao, Huanjian Zhou, Andi Han, Wei Huang, Masashi Sugiyama · 2 juillet 2026

  Discrete diffusion models are widely used for learning and generating discrete distributions. As the generation process is inherently sequential, the acceleration of sampling is of significant importance. In this work, we parallelize the mainstream $\tau$-leaping algorithm for absorbing discrete dif…

• Recovering Input Text from Hidden States: Study of Gradient-Based Inversion of Decoder-Only Language Models

  Miko{\l}aj S{\l}owikowski, Maciej Witold Majewski · 2 juillet 2026

  This work studies the hidden-state inversion problem: recovering the original input token sequence of a decoder-only language model from its last-layer hidden states. Rather than treating inversion as a one-shot reconstruction, we study it as a continuous embedding-space optimisation in which a soft…

• Flow-Map GRPO: Reinforcement Learning for Few-Step Flow-Map Generators via Anchored Stochastic Composition

  Zhiqi Li, Wen Zhang, Bo Zhu · 2 juillet 2026

  Few-step flow-map generators, such as consistency models and MeanFlow, accelerate sampling by directly learning long-range transport maps between noise and data. However, these models are typically deterministic, which makes them difficult to optimize with reinforcement learning (RL) post-training m…

• Controllable Narrative Rendering for Enhanced Assisted Writing

  Mingzhe Lu, Yanbing Liu, Jiayue Wu, Jiarui Zhang, Qihao Wang, Yue Hu, Yunpeng Li, Yangyan Xu · 2 juillet 2026

  Despite the remarkable proficiency of large language models (LLMs) in basic writing assistance, their utility in creative writing is fundamentally hindered by a persistent binary failure. This issue manifests as an oscillation between safe, surface-level editing, referred to as remedial polishing, a…

• Distilling Temporal Coherence into 2D Networks for Transrectal Ultrasound Prostate Video Segmentation

  Dong Yeong Kim, JunGyu Lee, Jaewon Choi, June Young Seo, Myeongseop Kim, Jinwook Choi, Taek Min Kim, Young-Gon Kim · 1 juillet 2026

  Real-time video segmentation of the prostate in Transrectal Ultrasound (TRUS) is essential for image-guided interventions. While conventional 2D methods suffer from inter-frame inconsistencies by disregarding temporal context, 3D architectures incur prohibitive latency. To resolve this dilemma, we p…

• Explaining Machine Learning and Memorization with Statistical Mechanics

  Robin Theriault · 1 juillet 2026

  Artificial neural networks (NNs) and machine learning (ML) algorithms are poorly understood from a theoretical perspective, which makes it difficult to fully realize their potential and overcome their weaknesses. For instance, ML algorithms train NN weights by moving them along a low-dimensional sub…

• SwiftAudio: Data-Efficient Caption-Only Distillation for One-Step Text-to-Audio Diffusion-based Generation

  Binh Mai, Tran Quoc Bao Le, Hung Dinh, Cong Tran · 1 juillet 2026

  Diffusion-based text-to-audio (TTA) models achieve impressive synthesis quality but suffer from high inference latency due to iterative multi-step denoising. Existing one-step approaches alleviate this issue but still rely on paired text--audio data during distillation. To address these limitations,…