Generative Adversarial Networks

Also known as GAN

Generative Adversarial Networks learn to produce data by training a generator against a discriminator that distinguishes generated samples from real ones. Common approaches include conditional GANs for controlled generation, Wasserstein GANs for more stable training, cycle-consistent models for unpaired translation, and StyleGAN-family models for high-quality images. GANs are used for image and audio synthesis, data augmentation, time-series modeling, and scientific applications, often with added constraints from labels, physics, or domain knowledge. A central tension is balancing realism, diversity, and training stability; strong visual or distributional metrics do not always predict whether synthetic data helps downstream tasks, and diffusion models are an important alternative.

Latest papers

178 papers20 papers in the last four weeks, up 18% on the four weeks before. 0.1% of all new papers.

The Search for Statistical Indistinguishability Between Real and Fake Images
Oct 28, 2026Mohaimin Rafi, Otavio Moraes Rosa, Anshuman Chhabra +1Image Generation Evaluation AI-Generated Image Detection
University of South Florida
ALADDIN: Adversarial Latent Diffusion Distillation for One-Step Identity Swapping
Oct 27, 2026Ivan Shchekotov, Timofei Gritsaev, Viacheslav Pirogov +1Image Generation Diffusion Model Distillation
Constructor University, Bremen, Germany · Sumsub, Berlin, Germany
Condition-Enhanced Generative Adversarial Network for Aerodynamic Configuration Design of Commercial Aircraft
Oct 25, 2026Youzheng Yu, Jiaqing Kou, Weiwei ZhangConditional Generative Modeling Generative Adversarial Networks
School of Aeronautics, Northwestern Polytechnical University, Xi’an, 710072, People’s Republic of China · International Joint Institute of Artificial Intelligence on Fluid Mechanics, Northwestern Polytechnical University, Xi’an, 710072, People’s Republic of China · National Key Laboratory of Aircraft Configuration Design, Xi’an, 710072, People’s Republic of China
L0-Norm-Based CycleGAN for Unpaired Brain CT-to-MR Synthesis
Oct 23, 2026Yunxiao Zhang, Shouqiang Du, Zexian Xu +1Medical Image Synthesis Unpaired Image-to-Image Translation
School of Mathematics and Statistics, Qingdao University, Qingdao, 266071, China. · The Affiliated Hospital of Qingdao University, Qingdao, 266003, China. · School of Stomatology, Qingdao University, Qingdao, 266003, China.
Forging Emotions: a deep learning experiment on emotions and art
Oct 23, 2026Amalia FokaImage Generation Affective Computing
University of Ioannina
Adversarial Post-Training of Foundation Models for High-Frequency Trading
Oct 19, 2026Alexandre Bismuth, Kang Li, Aramis Marti-Shahandeh +3Limit Order Book Modeling Adversarial Training
British Open-ended Learning and Discovery Lab (BOLD), Department of Computer Science, University of Oxford, Oxford, United Kingdom · Harvard University, Cambridge, MA, United States · British Open-ended Learning and Discovery Lab (BOLD), Department of Statistics, University of Oxford, Oxford, United Kingdom +1
Forecasting Voltage Time Series under Anomalies Using a WGAN Model With Lipschitz Control and Progressive Kernel Critic Strategy
Oct 16, 2026Najmieh Sadat SafarabadiTime Series Forecasting Robust Time Series Forecasting
Institut für Maschinelle Sprachverarbeitung, University Of Stuttgart, Stuttgart, Germany
Generative adversarial networks with Monte Carlo dropout uncertainty quantification for structural health monitoring: Fibre Bragg grating data augmentation comparison in Hawk T1A aircraft
Oct 15, 2026Alvaro Peiro, Marios ImpraimakisSynthetic Data Augmentation Synthetic Data Generation
Department of Mechanical Engineering, University of Bath, Bath, BA2 7AY, UK
Deep Learning for Atmospheric Turbulence Mitigation: A Comparison Between GANs and Transformers
Oct 13, 2026Juan S. Carrillo, Jérôme GillesAtmospheric Turbulence Mitigation Image Restoration
San Diego State University, Computational Science Research Center, Department of Mathematics and Statistics, 5500 Campanile Drive, San Diego, CA 92182, USA · San Diego State University, Department of Mathematics and Statistics, 5500 Campanile Drive, San Diego, CA 92182, USA
Weak Segmentation-Guided GAN for~Realistic Color Edition
Oct 12, 2026Vincent Auriau, Emmanuel Malherbe, Matthieu PerrotImage Editing Generative Adversarial Networks
L’Or´eal AI Research, France · Artefact Research Center, France
ProGAN-Based Synthetic Image Generation for Industrial Visual Inspection of Metallic Parts
Oct 12, 2026Hossein Omid Beiki, Sepehr Asgari, Oliver Petrovic +1Synthetic Data Generation Synthetic Data Augmentation
Laboratory for Machine Tools and Production Engineering WZL of RWTH, Aachen University, Aachen, Germany
Controlled-Minimax: Stabilizing Adversarial Optimization via Control Theory
Oct 9, 2026Tianjiao Luo, Jianfei Chen, Jun ZhuMin-Max Optimization Optimization Convergence Analysis
Department of Computer Science and Technology, Tsinghua University, Beijing, China.
Generative Adversarial Loops
Oct 8, 2026Kislay Aditya Oj, Nidhi Jain, Sri Surya Varma Datla +3Automated Algorithm Discovery Generative Adversarial Networks
Computer Science and Engineering, Indian Institute of Technology Bombay, Mumbai, India · Independent researcher · Electrical Engineering and Computer Sciences, University of California, Berkeley, USA
No Distillation Needed: Single-Pass Real-Time Talking Heads via Acausal Noise Shaping
Oct 8, 2026Yu Han, Dejan Markovic, Alexander Richard +4Talking Head Generation Audio-Driven Facial Animation
Meta Reality Labs
Adaptive Multi-Discriminator WGAN Framework for Resource-Constrained Internet of Vehicles Using Reinforcement Learning and Game Theory
Oct 7, 2026Farhoud Jafari Kaleibar, Amr M. Zaki, Marin LitoiuGenerative Adversarial Networks Intelligent Transportation Systems
Department of Electrical and Computer Engineering, York University, Toronto, ON, Canada
Gen-PINNs: Generative Adversarial Physics Informed Neural Networks for solving partial differential equations
Oct 7, 2026Muhammad M. Akmal, Kamy Sepehrnoori, Michael J. PyrczPDE Solving Physics-Informed ML
Hilderbrand Department of Petroleum and Geosystems Engineering, The University of Texas at Austin · Department of Earth and Planetary Sciences, The University of Texas at Austin
SwinDS-BWE: A Parameter-Efficient Swin-1D Lattice Generator with Decision-Science Discriminators for Speech Bandwidth Extension
Oct 6, 2026Tarikul Islam Tamiti, Sajid Fardin Dipto, Luke Baja-Ricketts +2Speech Enhancement Generative Adversarial Networks
Department of Cyber Security Engineering, George Mason University, Virginia, USA
Reading the Mood: Emotion-Guided Book-to-Music Recommendation via CGANs and LLMs
Oct 5, 2026Manousos Linardakis, Georgios AlexandridisCross-Domain Recommendation Recommender Systems
School of Electrical & Computer Engineering National Technical University of Athens Zografou, 15780, Greece · Department of Digital Industry Technologies National & Kapodistrian University of Athens Evripos Complex, Psachna, 34400, Greece
Multitask Conditional Generative Adversarial Network Enables Automatic Whole Knee Cartilage and Menisci Segmentation and Reliable $T_{1ρ}$ and $T_2$ Quantification Without High-Resolution Morphological Images
Oct 5, 2026Ahmed Tahseen Minhaz, Richard Lartey, Zhiyuan Zhang +8Multi-Task Learning Magnetic Resonance Imaging
Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic · Department of Biomedical Engineering, Cleveland Clinic Research, Cleveland Clinic · Department of Mathematics, Case Western Reserve University +1
EndoLive: Real-Time Style Transfer for Endoscopic Endonasal Skull Base Surgical Video
Oct 1, 2026Griffin Hurt, Calvin BrinkmanImage Style Transfer Unpaired Image-to-Image Translation
University of Pittsburgh School of Computing and Information
Parameterization method of reservoir properties for ensemble-based data assimilation using intermediate latent space of StyleGAN
Sep 30, 2026Marcio A. Sampaio, Paulo H. Ranazzi, Martin J. BluntData Assimilation Generative Modeling
Departamento de Engenharia de Minas e de Petróleo, Escola Politécnica, Universidade de São Paulo, 05508-030, SP, Brasil. · Department of Earth Science and Engineering, Imperial College London, South Kensington, London, UK
UDSS-BWE: Uncertainty- and Decision-Science Inspired Swin BandWidth Extension
Sep 28, 2026Tarikul Islam Tamiti, Sajid Fardin Dipto, David Vergano +2Generative Adversarial Networks
Department of Cyber Security Engineering, George Mason University, USA.
Marginal Calibration Masks Generator Differences in Held-Out Evaluation of Multivariate Time Series, with an Application to ERA5 Weather Sequences
Sep 27, 2026Mustafa Özaytaç, Özge Karadağ AtaşTime Series Generation Model Calibration
Department of Statistics, Graduate School of Science and Engineering, Hacettepe University, Ankara, 06800, Turkey · Department of Statistics, Faculty of Science, Hacettepe University, Ankara, 06800, Turkey
AnchorMixGAN: Anchor-Aligned Generative Semi-Supervision for DDoS Detection in Cloud-Integrated IoT Networks
Sep 26, 2026Jin Yang, Xufeng Liu, Yong Hu +3Network Intrusion Detection Cybersecurity
School of Cyber Science and Engineering, Sichuan University, Chengdu, 610207, China · College of Information Science and Technology, Xizang University, Lhasa, 850000, China · Shaanxi Key Laboratory of Intelligent Policing, Shaanxi Police College, Xi’an, 710021, China +2
EndoFSA: Endoscopic Few-Shot Image Generation via Rank-Constrained Parameter Adaptation
Sep 24, 2026Panagiota Gatoula, Grigoris Karypidis, Dimitris K. IakovidisImage Generation Medical Image Synthesis
Dept. of Computer Science and Biomedical Informatics University of Thessaly Lamia, Greece
Does DCGAN-Based Synthetic Augmentation Improve Brain Tumor MRI Classification? An Empirical Study
Sep 20, 2026Irhum Jawad Khan, Talha bin AslamSynthetic Data Augmentation Medical Image Classification
School of Electrical Engineering and Computer Science (SEECS) National University of Sciences and Technology (NUST) Islamabad, Pakistan · Independent Researcher Islamabad, Pakistan
When Does Adversarial Refinement Help? A Negative Result and Open Problem in Adapting R3GAN to Time Series Imputation
Sep 19, 2026Yufeng HeTime Series Imputation Generative Adversarial Networks
The University of Hong Kong Hong Kong, China
EmbeddGAN: A Novel GAN Framework Using an Embedding Network and Gini Distance Correlation
Sep 18, 2026MaTais Caldwell, Yixin Chen, Xin Dang +1Generative Adversarial Networks Adversarial Training
Department of Computer and Information Science University of Mississippi Oxford, MS 38677 USA · Department of Mathematics University of Mississippi Oxford, MS 38677 USA
HMB-GAN: Hybrid Multi-Bézier GAN for Vector Shape Synthesis
Sep 17, 2026Elian Hugh Thiele-Evans, Binh Duong Pham, Hani Omar M Alharbi +5Generative Adversarial Networks Quantum Generative Modeling
School of Science, Computing and Emerging Technologies Swinburne University of Technology Melbourne, Australia · School of Engineering, and School of Design and Architecture Swinburne University of Technology Melbourne, Australia
Beyond Pixel Similarity: Task-Aware Evaluation of GAN-Based Synthetic Sonar Data for Robotic Perception
Sep 16, 2026Hannan Ejaz Keen, Muhammad Moazam Fraz, Karsten BernsSynthetic Data Generation for Robotics Robotic Perception
XITASO GmbH IT & Software Solutions, Augsburg, Germany · School of Electrical Engineering and Computer Science at National University of Sciences and Technology (NUST), Islamabad, Pakistan · Department of Computer Science at Technical University of Kaiserslautern, Kaiserslautern, Germany

Oct 28, 2026J/K move · Enter open · S save

The Search for Statistical Indistinguishability Between Real and Fake Images

Mohaimin Rafi, Otavio Moraes Rosa, Anshuman Chhabra, Utkarsh Ojha

University of South Florida

Image Generation Evaluation AI-Generated Image Detection Image Generation Generative Adversarial Networks

Open paper AskPDF arXiv

Images produced by generative models appear strikingly similar to real ones. In this work, we study whether perfection has been reached by generated content in terms of matching the properties of real data. We train deep neural networks to separate real data from fake and observe their test accuracies. Only a chance performance has any scope of proving statistical indistinguishability. However, since generative models are known to sometimes introduce small errors, e.g., low level checkerboard artifacts, we create progressive abstractions of the original real/fake images to potentially bring them to a space where those artifacts get wiped away. These include operations such as blurring, downsizing, cropping, converting RGB images into edge/semantic maps, etc. To our surprise, almost none of the abstractions are able to make fake data indistinguishable from real. The only cases when neural network based classifiers fail to separate the two categories is when the data is abstracted to such an extent that the original content seems mostly lost to the human eye, e.g., downsizing a

256 \times 256

RGB image of human face to

4 \times 4

resolution. We observe this trend across four different datasets of varying complexity (MNIST, CelebA, FFHQ and GenImage) and across various types of generative models (GANs, VAEs, diffusion models). This shows that the nature of fake artifacts extends all the way from low level to very high level in images. The simple implication is that fake images are much more different from real than we are used to think.