mlsanityData to 7 Oct 2026

NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild

arXiv 2604.11487 ↗First posted 13 Apr 2026cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

This paper presents an overview of the NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild, held in conjunction with the NTIRE workshop at CVPR 2026. The goal of this challenge was to develop detection models capable of distinguishing real images from generated ones in realistic scenarios: the images are often transformed (cropped, resized, compressed, blurred) for practical usage, and therefore, the detection models should be robust to such transformations. The challenge is based on a novel dataset consisting of 108,750 real and 185,750 AI-generated images from 42 generators comprising a large variety of open-source and closed-source models of various architectures, augmented with 36 image transformations. Methods were evaluated using ROC AUC on the full test set, including both transformed and untransformed images. A total of 511 participants registered, with 20 teams submitting valid final solutions. This report provides a comprehensive overview of the challenge, describes the proposed solutions, and can be used as a valuable reference for researchers and practitioners in increasing the robustness of the detection models to real-world transformations.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 49 of its 77 references that are in the map. First posted on arXiv 13 Apr 2026.

Builds on

49 of 77 references.

References · 49

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  6. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  7. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  8. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  9. 2022Generative Adversarial Networkscited by 5,603
  10. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  11. 2022A ConvNet for the 2020scited by 3,545
  12. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
37 more
  1. 2020Supervised Contrastive Learningcited by 2,803
  2. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,360
  3. 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,393
  4. 2022CoCa: Contrastive Captioners are Image-Text Foundation Modelscited by 1,265
  5. 2021Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Conceptscited by 1,209
  6. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
  7. 2025SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Featurescited by 1,127
  8. 2025Qwen-Image Technical Reportcited by 973
  9. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Spacecited by 960
  10. 2019CNN-generated images are surprisingly easy to spot... for nowcited by 842
  11. 2023EVA-CLIP: Improved Training Techniques for CLIP at Scalecited by 786
  12. 2021Robust fine-tuning of zero-shot modelscited by 785
  13. 2024Autoregressive Image Generation without Vector Quantizationcited by 644
  14. 2023DataComp: In search of the next generation of multimodal datasetscited by 635
  15. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 570
  16. 2020Leveraging Frequency Analysis for Deep Fake Image Recognitioncited by 501
  17. 2023Towards Universal Fake Image Detectors that Generalize Across Generative Modelscited by 483
  18. 2023DIRE for Diffusion-Generated Image Detectioncited by 386
  19. 2023EVA-02: A Visual Representation for Neon Genesiscited by 377
  20. 2023GenImage: A Million-Scale Benchmark for Detecting AI-Generated Imagecited by 318
  21. 2022On the detection of synthetic images generated by diffusion modelscited by 282
  22. 2024A Sanity Check for AI-generated Image Detectioncited by 187
  23. 2021RedCaps: web-curated image-text data created by the people, for the peoplecited by 175
  24. 2021Are GAN generated images easy to detect? A critical analysis of the state-of-the-artcited by 139
  25. 2025HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformercited by 131
  26. 2024DF40: Toward Next-Generation Deepfake Detectioncited by 126
  27. 2023ARNIQA: Learning Distortion Manifold for Image Quality Assessmentcited by 87
  28. 2026The Eleventh NTIRE 2026 Efficient Super-Resolution Challenge Reportcited by 25
  29. 2025Bridging the Gap Between Ideal and Real-world Evaluation: Benchmarking AI-Generated Image Detection in Challenging Scenarioscited by 24
  30. 2026NTIRE 2026 The 3rd Restore Any Image Model (RAIM) Challenge: Multi-Exposure Image Fusion in Dynamic Scenes (Track 2)cited by 16
  31. 2026NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Resultscited by 15
  32. 2026NTIRE 2026 Challenge on Single Image Reflection Removal in the Wild: Datasets, Results, and Methodscited by 15
  33. 2026NTIRE 2026 Challenge on Short-form UGC Video Restoration in the Wild with Generative Models: Datasets, Methods and Resultscited by 14
  34. 2026NTIRE 2026 The Second Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Resultscited by 14
  35. 2026The Second Challenge on Real-World Face Restoration at NTIRE 2026: Methods and Resultscited by 13
  36. 2025Transferable Black-Box One-Shot Forging of Watermarks via Image Preference Modelscited by 10
  37. 2025Ovis-Image Technical Reportcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 18.

18 papers.

Sort:

By area:

  1. 2026NTIRE 2026 The 3rd Restore Any Image Model (RAIM) Challenge: Professional Image Quality Assessment (Track 1)cited by 12
  2. 2026The Fourth Challenge on Image Super-Resolution (×4) at NTIRE 2026: Benchmark Results and Method Overviewcited by 10
  3. 2026NTIRE 2026 Challenge on Video Saliency Prediction: Methods and Resultscited by 10
  4. 2026NTIRE 2026 Rip Current Detection and Segmentation (RipDetSeg) Challenge Reportcited by 8
  5. 2026Low Light Image Enhancement Challenge at NTIRE 2026cited by 7
  6. 2026The First Challenge on Mobile Real-World Image Super-Resolution at NTIRE 2026: Benchmark Results and Method Overviewcited by 6
  7. 2026The First Challenge on Remote Sensing Infrared Image Super-Resolution at NTIRE 2026: Benchmark Results and Method Overviewcited by 6
  8. 2026Robust Deepfake Detection, NTIRE 2026 Challenge: Reportcited by 6
  9. 2026The First Controllable Bokeh Rendering Challenge at NTIRE 2026cited by 2
  10. 2026The Third Challenge on Image Denoising at NTIRE 2026: Methods and Resultscited by 1
  11. 2026Boosting Robust AIGI Detection with LoRA-based Pairwise Trainingcited by 0
  12. 2026NTIRE 2026 Challenge on Efficient Low Light Image Enhancement: Methods and Resultscited by 0
6 more
  1. 2026Pixel-Space Diffusion Transformerscited by 0
  2. 2026NTIRE 2026 Low-light Enhancement: Twilight Cowboy Challengecited by 0
  3. 2026LAION-Mobile: Evaluating Deepfake Detectors On One Million Smartphone Photoscited by 0
  4. 2026REALIS: A Curated Dataset for Studying the Challenges of AI Image Detectioncited by 0
  5. 2026Mandela-Bench: Multimodal Models Remember Canonical Images Instead of Seeing Themcited by 0
  6. 2026Trustworthy synthetic visual media: Evidence across the media lifecyclecited by 0