Generative Adversarial Networks and Image Synthesis

Derniers papiers

• Diffusion models recover accurate mixture weights despite score function insensitivity

  Andrew Dennehy, Ramchandran Muthukumar, Rebecca Willett, Nisha Chandramoorthy · 20 juillet 2026

  Score-based generative models exhibit a puzzling behavior: they often appear to cover all modes of a target multimodal distribution and yet may fail to learn the correct relative mode amplitudes, which can be interpreted as mixture weights. We resolve this apparent paradox by relating the diffusion …

• Loop the Loopies!

  Zitian Gao, Yilong Chen, Yihao Xiao, Xinyu Yang, Ran Tao, Joey Zhou, Bryan Dai · 20 juillet 2026

  We present Loopie, the most powerful looped Transformer to date. The Loopie series consists of two Mixture-of-Experts (MoE) models: a 20B-parameter model with 2B active parameters and a 6Bparameter model with 0.6B active parameters. Looped Transformers have long faced a challenge: given an N-fold in…

• On the Failure of Boundary-Seeking Distillation in Bottlenecked Generative Architectures

  Mohamed Amine Kina · 20 juillet 2026

  Data-free knowledge distillation transfers the knowledge encoded in a teacher model to a student model without access to the original training data. Prior work such as Contrastive Abductive Knowledge Extraction (CAKE) achieves this for classifiers by synthesizing samples near the teacher's decision …

• Test-Time Noise Guided Adaptation for Realistic Autoregressive Video Generation

  Dimitrios Karageorgiou, Symeon Papadopoulos, Ioannis Kompatsiaris, Efstratios Gavves · 20 juillet 2026

  Autoregressive video diffusion models have enabled the generation of arbitrarily long videos by removing conditioning on future frames, thus greatly improving computational efficiency. Yet, they suffer from error accumulation over time, as the denoised sequence gradually drifts away from the conditi…

• Inference-Time Concept Suppression and Video-Centric Evaluation for Text-to-Video Models

  Wenxuan Chen, Wenjie Feng · 17 juillet 2026

  Text-to-video (T2V) generators can synthesize realistic and temporally coherent videos, but controllably removing a target concept from a generator remains difficult. Unlike text-to-image concept erasure, T2V unlearning must suppress a target concept that may persist across frames while preserving…

• Integration Matters: Rollout-Based Training for Constrained Diffusion Models

  Xiaoxuan Liang, Saeid Naderiparizi, Berend Zwartsenberg, Frank Wood · 17 juillet 2026

  Constrained generative models aim to produce samples that satisfy complex feasibility constraints while remaining faithful to the data distribution. Existing constrained generation methods typically enforce constraints either through training-time optimization or sampling-time correction. Training-t…

• Advanced Image Generation: Negative Prompt Optimization and Latent Classifier Guidance

  Vaddi Charan Sai Nandan Reddy, Harini B, Chandana M S · 17 juillet 2026

  We present a novel system that integrates negative prompt optimization via a fine-tuned sequence-to-sequence LLM and latent-space classifier guidance to improve the quality of images generated by Stable Diffusion. Our approach automatically generates optimized negative prompts, and employs a CNN-RNN…

• Generalized Neural Distributional Regression

  Natan Hilario da Silva, Vicente Garibay Cancho, Adriano Kamimura Suzuki · 17 juillet 2026

  We introduce the Generalized Neural Distributional Regression (GNDR) framework, which seamlessly embeds deep neural networks into the parameter space of classical probability distributions. To reconcile the inherent non-identifiability of deep architectures with maximum likelihood theory, we propose…

• Multi-Axis Max@K Reinforcement Learning for Representative Diversity in Text-to-Image Generation

  Ku Onoda, Paavo Parmas, Hiroki Furuta, Soichiro Nishimori, Yuta Oshima, Shohei Taniguchi, Yutaka Matsuo · 17 juillet 2026

  Text-to-image (T2I) models can synthesize realistic, prompt-aligned images, yet samples generated for the same prompt often cover only a small subset of visually distinct modes. This limits the diversity of images, and for person-centric prompts, can reflect or amplify demographic skew. We formalize…

• Continuously Evolving Deepfake Detection: An Architecture and Public-Benchmark Evaluation of a Dynamic Detection System

  Ken Jon Miyachi, Dylan Uys · 16 juillet 2026

  Deepfake detectors that achieve near-perfect scores on academic benchmarks collapse on real-world content: recent in-the-wild evaluations report AUC drops of 45-50% for state-of-the-art open-source models. We argue this gap is structural: static detectors are trained once against a moving generative…

• Hallo4D: Multi-Modal Hallucination Mitigation for Consistent Spatio-Temporal Generation

  Hongbo Wang, Huaibo Huang, Jie Cao, Jin Liu, Haoyang Tong, Ran He · 16 juillet 2026

  While recent advances in 3D generation have enabled impressive visual synthesis, existing methods often rely on 2D diffusion supervision without explicit mechanisms for geometric consistency, leading to spatial hallucinations such as duplicated structures and misaligned geometry. These issues become…

• Targeted Recovery of Weight-Space Mechanisms From Neural Networks

  Antoine Vigouroux, Lee Sharkey · 16 juillet 2026

  Parameter decomposition (PD) decomposes neural networks into interpretable computational components that faithfully reflect the original network's operations. However, scaling PD to large models requires vast compute, making it a costly and risky endeavor. Here we propose targeted PD (tPD), which id…

• Kaleido: Algorithm-Hardware Co-Design for Video Diffusion Transformers by Exploiting Latent Space Correlations

  Wenxuan Miao, Haosong Liu, Weiming Hu, Zihan Liu, Aiyue Chen, Jianlin Yu, Yiwu Yao, Yiming Gan, Jieru Zhao, Jingwen Leng, Minyi Guo, Yu Feng · 16 juillet 2026

  Video diffusion transformers (vDiTs) generate high quality video but introduce extremely high compute cost due to the long diffusion timesteps and self attention computation. As diffusion timesteps are reduced, the computation cost of self attention becomes the dominant bottleneck. Existing accelera…

• Fre-Res: Frequency-Residual Video Token Compression for Efficient Video MLLMs

  Yigui Feng (The College of Computer Science, National University of Defense Technology, Changsha, Hunan, China), Qinglin Wang (The College of Computer Science, National University of Defense Technology, Changsha, Hunan, China), Yang Liu (The Shien-Ming Wu School of Intelligent Engineering, South China University of Technology, Guangzhou, Guangdong, China), Jie Liu (The College of Computer Science, National University of Defense Technology, Changsha, Hunan, China) · 16 juillet 2026

  Video MLLMs face a persistent tension between spatial fidelity and temporal coverage: preserving fine-grained visual details requires many spatial tokens, while capturing short-lived events requires dense temporal sampling. We propose \textbf{Fre-Res}, a budget-adaptive dual-track video-token compre…

• HIVE-3D: Hierarchical Voxel Enhancement for High-Quality 3D Scene Generation

  Bin Zang, Wenting Zheng, Xiaoliang Luo, Zhiyuan Fang, Shi Li, Lvchun Wang, Wei Yu, Yi Zhao, Tian Xie, Yuchi Huo, Rengan Xie · 16 juillet 2026

  Recently, a line of works can generate impressive 3D objects from a single image, but they are limited by restricted representation resolution, making them unsuitable for 3D scene generation. In this work, we introduce HIVE-3D, a novel method for high-quality 3D scene generation based on hierarchica…

• The SIGReg Objective as Variational Free Energy: A Theoretical Active-Inference Account of JEPA World Models

  Fabio Arnez, Alexandra Gomez-Villa · 16 juillet 2026

  Joint-Embedding Predictive Architectures (JEPAs) are the dominant design for latent world models, yet they are usually justified by empirical performance rather than a normative principle. We show that the choice of anti-collapse regulariser determines whether a JEPA's training objective, a predicti…

• Concurrent Image Understanding and Generation: Self-Correcting Coupled Markov Jump Processes

  Minh-Quan Le, Armand Comas, Alexandros Lattas, Stylianos Moschoglou, Pedro V\'elez, Amit Raj, Aaron Germuth, Thabo Beeler, Dimitris Samaras, Di Qiu · 16 juillet 2026

  Human cognition does not separate understanding and generation. A teacher at a whiteboard speaks and draws $\textit{together}$, each modality reshapes the other. In this paper, we bring this coupled loop to artificial systems. Masked Diffusion Models (MDMs) are ideally suited to this task, yet exist…

• Self-Improving is Often Sudden: Enlightenment-style Finetuning for Large-Scale Models

  Jing-Xiao Liao, Tianwei Zhang, Yu-Hao Jiang, Feifei Zhang, Hang-Cheng Dong, Feng-Lei Fan · 16 juillet 2026

  The pursuit of autonomously self-improving models has attracted growing interest in the era of large-scale foundation models. Drawing inspiration from the concept of "enlightenment" or "aha moment" in human brain, we hypothesize that large models exhibit an analogous enlightenment phenomenon-a laten…

• Higher Embedding Dimension Creates a Stronger World Model for a Simple Sorting Task

  Brady Bhalla, Honglu Fan, Nancy Chen, Tony Yue YU · 15 juillet 2026

  We investigate how embedding dimension affects the emergence of an internal "world model" in a transformer trained with reinforcement learning to perform bubble-sort-style adjacent swaps. Models achieve high accuracy even with very small embedding dimensions, but larger dimensions yield more faithfu…

• From Geometric Recovery to Causal Validation: A Reproducible Audit of Sparse Autoencoder Features, from Superposition Geometry to Causal Inertness

  Mohamed Abdessalem Bal · 15 juillet 2026

  Sparse autoencoders (SAEs) are the standard for decomposing superposed neural representations into interpretable features, and evaluation relies predominantly on correlational recovery metrics -- cosine similarity between ground-truth directions and decoder atoms. We show this conflates two distinct…

• FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model Serving

  Yaqi Qiao, Ping He, Songrun Xie, Ayush Barik, Chensong Zhang, Zhengzhong Tu, Fan Lai · 15 juillet 2026

  Diffusion models have become the central backbone for modern image, video, and audio generation, but their efficient service remains a challenge. Unlike autoregressive decoding, diffusion inference repeatedly updates high-dimensional spatial or temporal latents over many denoising steps. This all-re…

• The Geometry of Memorization: Finite-Time Spectral Sensitivity as a Diagnostic for Flow Matching Models

  Shuchan Wang · 15 juillet 2026

  Continuous-time generative frameworks construct probability paths between base and target domains by optimizing time-dependent velocity fields. While theoretical targets favor straight trajectories, empirical networks develop complex path deformations. This paper presents the Finite-Time Spectral Se…

• Do You Remember? Toward Memory-Centric Multimodal AI

  Xuguang Yu, Weigang Zheng, Minyue Yu · 15 juillet 2026

  Human memory is reconstructive, not a faithful recording. Current multimodal LLMs (MLLMs) lack this capability: they process images through a frozen visual encoder, produce a one-shot text output, and discard internal representations. We present DoYouRemember, a three-stage architecture introducing …

• Self-Consistent Flow: Unifying Velocity and Endpoint Prediction for Rectified Flow Models

  Xu Han, Jiajing Hu, Li-Ping Liu · 15 juillet 2026

  In rectified-flow-based generative models, the neural network can be trained to predict two different targets, such as the instantaneous velocity or the data endpoint, to perform denoising. Although prior work shows that these parameterizations lead to different empirical behaviors, the mechanisms u…

• PolarBM: Complex-valued Boltzmann Machine for Modeling Audio Signals in Polar and Log-polar Coordinates

  Toru Nakashika, Kohei Yatabe · 15 juillet 2026

  Although vast amounts of data, such as audio signal spectra, are naturally represented using complex numbers, conventional machine learning methods often simplify complex-domain problems by employing frameworks designed for real-valued variables. While this simplification offers computational benefi…