REVIEW 3 major objections 6 minor 47 references
MLEP: Multi-granularity Local Entropy Patterns for Universal AI-generated Image Detection
T0 review · 3 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash
Pith's one-line read Local entropy patterns spot AI images across 32 generators
desk verdict A plausible, cheap entropy-based feature for AIGI detection with a real but possibly inflated mean gain: the headline 97.1% is selected from ablations on the same test datasets, with no held-out validation or error bars. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the MLEP feature map. Construction: split an image into $2\times2$ patches, shuffle them randomly (per color channel), downsample and upsample the scrambled image at scales $S=\{1,1/2,1/4\}$ with bilinear interpolation, then slide a $2\times2$ window with stride 1 over every resampled version and record the Shannon entropy of the four pixels inside each window. Because only four values enter the entropy formula, the LEP output takes values in $\{0,0.8,1.0,1.5,2.0\}$, and the overlapping windows capture intra-patch, inter-patch, and inter-scale randomness at once. Patch shuffling destroys the semantic layout, multi-scale resampling exposes upsampling artifacts common to generators, and the entropy statistic replaces pixel contrasts with a measure of local randomness. The resulting multi-channel maps are concatenated and fed to a standard ResNet-50 trained with binary cross-entropy.
What would settle it
Fix hyperparameters ($l=2$, $S=\{1,1/2,1/4\}$, stride 1, bilinear interpolation) using a held-out validation set of a few generators, then test on a generator released after training that is not among the 32, or apply mild JPEG compression and Gaussian blur to the existing test sets. If mean accuracy drops to baseline levels, the claim of a universal entropy fingerprint fails.
Extended reading notes
Core claim
The paper's central discovery is that local entropy distributions over shuffled micro-patches separate real photographs from AI-generated images in a way that transfers across generator families. A real image consistently has a higher probability of maximum entropy (2.0) in 2x2 windows than a generated one, which the authors attribute to the glossy, over-smoothed texture generators produce. By computing entropy on randomly permuted $2\times2$ patches across the resampling scales $S=\{1,1/2,1/4\}$, MLEP suppresses scene semantics while preserving and amplifying generator-specific randomness. Trained only on ProGAN images from ForenSynths and real LSUN images, a ResNet-50 classifier on MLEP maps achieves 97.1% mean accuracy over 32 generation models, with gains of 3.1% in accuracy and 1.6% in average precision over the nearest strong baseline NPR. The authors frame this as the first use of image entropy as a detection cue for AI-generated images.
Load-bearing premise
The entire reported advantage rests on the assumption that the hyperparameters were not effectively tuned to the 32 test datasets; the paper selects patch size, scales, stride, and interpolation through ablations on those same datasets without describing a held-out validation split for model selection.
Editorial extensions
If this is right
- A detector trained only on ProGAN images transfers to 16 GAN and 16 diffusion models, reaching 97.1% mean accuracy across all 32.
- Patch size $l=2$ beats $l=4$ and $l=8$, supporting the claim that finer semantic scrambling improves content-agnostic detection.
- Entropy alone, without shuffling or multi-scale resampling, already exceeds the NPR baseline, indicating the entropy statistic itself carries the signal.
- Multi-scale fusion with scales $\{1,1/2,1/4\}$ is optimal; adding $1/8$ hurts, so moderate resampling introduces useful artifacts without excessive distortion.
- Performance is stable across ResNet-18, ResNet-34, ResNet-50, and ResNet-101, so the representation rather than a particular classifier depth drives the result.
Reading between the lines
- One testable extension: evaluate MLEP under JPEG compression, blur, and noise; the paper's own limitation section says these are unstudied, and a drop there would bound its real-world use.
- Because the feature is a five-value entropy pattern, a fast implementation could classify images in real time on edge devices; the paper does not measure latency or throughput.
- If over-smoothed texture is the mechanism, MLEP should also catch partially generated edits such as face swaps and inpainting, a benchmark the paper does not include.
- The claim of content agnosticism could be stress-tested by training on LSUN scenes and testing on face or document image datasets, where texture statistics differ.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes Multi-granularity Local Entropy Patterns (MLEP), a handcrafted feature representation for AI-generated image detection. The method computes Shannon entropy over 2x2 sliding windows on randomly shuffled small patches, repeats this at multiple resampling scales, and feeds the resulting multi-channel entropy maps to a ResNet classifier trained only on ProGAN images. The authors evaluate on 32 datasets from 16 GAN and 16 diffusion models and report a mean accuracy of 97.1%, surpassing the NPR baseline by 3.1 accuracy points. They also provide ablations on patch size, scale factors, interpolation method, stride, and backbone, plus qualitative visualizations of entropy maps and t-SNE features to support their semantic-suppression argument.
Significance. If the reported results are unbiased, MLEP is an interesting and practically simple contribution: it requires no pretrained generative model or vision-language backbone, it uses only a standard CNN on a derived entropy representation, and its cross-generator generalization (trained on ProGAN, tested on 32 generators) is strong. The paper also gives a clear qualitative story for why patch shuffling and local entropy should suppress semantic content. These strengths are real and make the method worth pursuing. However, the central quantitative claim — a 3.1-point mean accuracy advantage over NPR and a 5.2-point advantage over FatFormer — is currently supported by an evaluation protocol that selects hyperparameters on the same test datasets that produce the headline numbers, and it reports no error bars or significance tests. The claim of 'universal' detection is also weakened by per-dataset results where the method is far behind the baselines. The significance of the paper as written therefore rests on an evaluation that needs to be made statistically sound before the conclusions can be trusted.
major comments (3)
- [§4.1 and §4.2.2 (Tables 5–8)] The hyperparameters of MLEP — patch size l=2, scales S={1,1/2,1/4}, bilinear interpolation, stride 1, and the ResNet-50 backbone — are selected by maximizing mean accuracy/AP on the same 16 GAN and 16 diffusion test datasets that are later used for the headline results in Tables 1–3. The text does not describe any held-out validation split. This is selection on the test distribution and can inflate the reported 97.1% mean accuracy. It also makes the comparison to NPR and FatFormer unfair, since those baselines were not tuned on these 32 datasets and several were taken directly from the literature with their published settings. Please introduce a proper validation split (e.g., held-out categories of ForenSynths, or a separate set of generative models for validation), or report results for all hyperparameter combinations and discuss sensitivity explicitly.
- [§4.2.1 (Tables 1–3)] All reported numbers come from a single run, and the seed for the random patch shuffling is not reported. Because patch shuffling introduces randomness, the mean accuracy and AP could vary from run to run; the claimed advantage over NPR (97.1 vs 94.0) could be within run-to-run noise. Please report means and standard deviations over at least five runs for the key comparisons, and include a paired significance test (e.g., a paired t-test or Wilcoxon test over the 32 datasets) against the strongest baselines NPR and FatFormer.
- [§4.2.1, Tables 1 and 3] The paper concludes that MLEP 'outperforms state-of-the-art approaches by over 3% in accuracy' and calls the method 'universal', but this is based only on the mean over all 32 datasets. On the GAN subset, MLEP achieves 96.4 Acc versus FatFormer's 99.0 (Table 3), and on several individual GAN datasets the gap is much larger: BigGAN 87.1 vs 99.5 for FatFormer, GauGAN 82.0 vs 99.4, and S3GAN 83.4 vs 99.0 (Table 1). These are not isolated drops; they are on 3 of the 16 GAN datasets and contradict the impression of uniform superiority. The authors should report per-domain results, discuss the GAN-specific weakness, and temper the conclusion accordingly. A mean-only reporting is not sufficient to substantiate the 'universal' claim.
minor comments (6)
- [Abstract] "multiple image scaled" should be "multiple image scales".
- [§3.3] The set V={0, 0.8, 1.0, 1.5, 2} suggests that entropy values are rounded to one decimal; please define the rounding rule explicitly and state how ties in the four pixel values are handled in the probability estimate.
- [§3.3] The text says "The proof and an efficient computation algorithm for LEP on a 2×2 window are provided in the supplementary material." If the supplementary material is not part of this submission, the proof should be included in the main text or omitted; otherwise the statement is not verifiable.
- [Figure 1] The caption lists percentages (e.g., 45.6%, 43.9%) but does not explain what they represent; please clarify whether they are the fraction of pixels with entropy 2.0, or some other quantity.
- [References] Reference [1] has a typo: "Cramer Cistance" should be "Cramer Distance", and reference [27] has "Fiffusion" which should be "Diffusion".
- [§4.2.3] The t-SNE visualizations (Fig. 6) are qualitative; consider adding a quantitative separation metric (e.g., silhouette score or nearest-neighbor accuracy) so the claim of cleaner separation is testable.
Circularity Check
The headline 97.1% mean accuracy is selected on the same test datasets, so the reported 'prediction' is partly a fitted maximum rather than an unbiased generalization estimate.
-
fitted input called prediction
[Section 4.1 Implementation Details; Section 4.2.2 Table 5; final results in Section 4.2.1 Tables 1-3]
"To evaluate the effects of patch size (l) and resampling scales (S), we tested l∈{2, 4, 8} and S ={1, 1/2, 1/4, 1/8}. ... The best performance was achieved with the smallest patch size (l = 2), indicating that stronger semantic scrambling improves detection. ... We evaluated the generalization performance ... using patch size l = 2 , scales S = {1, 1/2, 1/4}, and a ResNet-50 backbone."
Tables 5-8 report Acc./A.P. on the same GAN-set, Diff.-set, and Mean collections that later produce the headline numbers in Tables 1-3. The final configuration (l=2, S={1,1/2,1/4}, stride 1, bilinear, ResNet-50) is chosen because it achieved the best performance in Table 5, then the same test datasets are used to report 97.1% mean accuracy. Thus the reported number is the maximum over a small hyperparameter grid evaluated with test labels, not the expected accuracy of a pre-specified method. The paper describes no held-out validation split, and baselines are not reported as equivalently tuned on these 32 datasets, so the 3.1-point margin over NPR is partly a selection artifact.
full rationale
The feature-construction part of the paper is not circular: MLEP is explicitly defined via patch shuffling, multi-scale resampling, and Shannon entropy on 2x2 windows, and the classifier is trained only on ProGAN images. There is no self-citation chain, no imported uniqueness theorem, and no renaming of a known result; the entropy descriptor is self-contained and the empirical comparison is meaningful in principle. The circularity is confined to the evaluation protocol: hyperparameters are selected using the same test datasets whose accuracy is then reported as the method's generalization performance. This makes the headline 97.1% figure, and the claimed >3% gain over NPR, partly constructed from the test labels rather than predicted. Because the central efficacy claim depends on that reported number, the paper earns a partial-circularity score of 6 rather than a clean bill of health. If the authors re-ran the protocol with a held-out validation split or pre-registered hyperparameters, the same method could receive a low circularity score.
Assumptions & free parameters
free parameters (4)
- patch_size_l =
2
- resampling_scales_S =
{1, 1/2, 1/4}
- sliding_window_stride =
1
- interpolation_method =
bilinear
assumptions (5)
- domain assumption Shannon entropy on a 2x2 window, with values confined to {0, 0.8, 1.0, 1.5, 2}, measures pixel randomness relevant to generation artifacts
- domain assumption Random shuffling of small patches removes semantic content without destroying the artifacts entropy is meant to capture
- domain assumption Downsampling and upsampling the scrambled image introduces resampling artifacts that differ between real and generated images
- domain assumption A ResNet-50 CNN trained on MLEP maps of ProGAN images can generalize to unseen GANs and diffusion models
- standard math Standard arithmetic and information theory definitions
Cite this review
Pith. "Pith review of MLEP: Multi-granularity Local Entropy Patterns for Universal AI-generated Image Detection." pith.science (2026). https://pith.science/paper/WUKY4Z5X
@misc{pith2026250413726,
author = {Pith},
title = {Pith review of: MLEP: Multi-granularity Local Entropy Patterns for Universal AI-generated Image Detection},
year = {2026},
howpublished = {\url{https://pith.science/paper/WUKY4Z5X}},
note = {Machine review of arXiv:2504.13726}
}
read the original abstract
Advancements in image generation technologies have raised significant concerns about their potential misuse, such as producing misinformation and deepfakes. Therefore, there is an urgent need for effective methods to detect AI-generated images (AIGI). Despite progress in AIGI detection, achieving reliable performance across diverse generation models and scenes remains challenging due to the lack of source-invariant features and limited generalization capabilities in existing methods. In this work, we explore the potential of using image entropy as a cue for AIGI detection and propose Multi-granularity Local Entropy Patterns (MLEP), a set of entropy feature maps computed across shuffled small patches over multiple image scaled. MLEP comprehensively captures pixel relationships across dimensions and scales while significantly disrupting image semantics, reducing potential content bias. Leveraging MLEP, a robust CNN-based classifier for AIGI detection can be trained. Extensive experiments conducted in an open-world scenario, evaluating images synthesized by 32 distinct generative models, demonstrate significant improvements over state-of-the-art methods in both accuracy and generalization.
Figures
Figures from the paper (3 more)
Reference graph
Works this paper leans on
-
[1]
The Cramer Cistance as a Solution to Biased Wasserstein Gradients
Marc G Bellemare, Ivo Danihelka, Will Dabney, Shakir Mohamed, Balaji Lakshminarayanan, Stephan Hoyer, and Rémi Munos. The Cramer Cistance as a Solution to Biased Wasserstein Gradients. In International Conference on Learning Representations, 2018
work page 2018
-
[2]
BEGAN: Boundary Equilibrium Generative Adversarial Networks
David Berthelot. BEGAN: Boundary Equilibrium Generative Adversarial Networks. arXiv preprint arXiv:1703.10717, 2017
arXiv 2017
-
[3]
Large Scale GAN Training for High Fidelity Natural Image Synthesis
Andrew Brock, Jeff Donahue, and Karen Simonyan. Large Scale GAN Training for High Fidelity Natural Image Synthesis. In International Conference on Learning Representations, 2018
work page 2018
-
[4]
Baoying Chen, Jishen Zeng, Jianquan Yang, and Rui Yang. DRCT: Diffusion Reconstruction Contrastive Training towards Universal Detection of Diffusion Generated Images. In Forty-first International Conference on Machine Learning, 2024
work page 2024
-
[5]
StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Transla- tion
Yunjey Choi, Minje Choi, Munyoung Kim, Jung-Woo Ha, Sunghun Kim, and Jaegul Choo. StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Transla- tion. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 8789–8797, 2018
work page 2018
-
[6]
Diffusion Models Beat GANs on Image Synthesis
Prafulla Dhariwal and Alexander Nichol. Diffusion Models Beat GANs on Image Synthesis. Advances in neural information processing systems, 34:8780–8794, 2021
work page 2021
-
[7]
Leveraging Frequency Analysis for Deep Fake Image Recognition
Joel Frank, Thorsten Eisenhofer, Lea Schönherr, Asja Fischer, Dorothea Kolossa, and Thorsten Holz. Leveraging Frequency Analysis for Deep Fake Image Recognition. In International Conference on Machine Learning, pages 3247–3258. PMLR, 2020
work page 2020
-
[8]
Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative Adversarial Nets. InProceedings of the 28th International Conference on Neural Information Processing Systems - Volume 2, NIPS’14, page 2672–2680, 2014
work page 2014
Show all 47 references
-
[9]
Vector Quantized Diffusion Model for Text-to-Image Synthesis
Shuyang Gu, Dong Chen, Jianmin Bao, Fang Wen, Bo Zhang, Dongdong Chen, Lu Yuan, and Baining Guo. Vector Quantized Diffusion Model for Text-to-Image Synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10696–10706, 2022
2022
-
[10]
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep Residual Learning for Image Recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, pages 770–778, 2016
2016
-
[11]
AttGAN: Facial Attribute Editing by Only Changing What You Want
Zhenliang He, Wangmeng Zuo, Meina Kan, Shiguang Shan, and Xilin Chen. AttGAN: Facial Attribute Editing by Only Changing What You Want. IEEE transactions on image processing, 28(11):5464–5478, 2019
2019
-
[12]
Denoising Diffusion Probabilistic Models.Advances in neural information processing systems, 33:6840–6851, 2020
Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising Diffusion Probabilistic Models.Advances in neural information processing systems, 33:6840–6851, 2020
2020
-
[13]
This isn’t the Nature Podcast-how deepfakes are distorting reality
Nick Petri´c Howe and Benjamin Thompson. This isn’t the Nature Podcast-how deepfakes are distorting reality. Nature, 2023
2023
-
[14]
Progressive Growing of Gans for Improved Quality, Stability, and Variation
Tero Karras, Timo Aila, Samuli Laine, and Jaakko Lehtinen. Progressive Growing of Gans for Improved Quality, Stability, and Variation. In International Conference on Learning Representations, 2018
2018
-
[15]
A Style-Based Generator Architecture for Genera- tive Adversarial Networks
Tero Karras, Samuli Laine, and Timo Aila. A Style-Based Generator Architecture for Genera- tive Adversarial Networks. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4396–4405, 2019. 10
2019
-
[16]
Analyzing and Improving the Image Quality of StyleGAN
Tero Karras, Samuli Laine, Miika Aittala, Janne Hellsten, Jaakko Lehtinen, and Timo Aila. Analyzing and Improving the Image Quality of StyleGAN. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8110–8119, 2020
2020
-
[17]
CLIPping the Deception: Adapting Vision- Language Models for Universal deepfake detection
Sohail Ahmed Khan and Duc-Tien Dang-Nguyen. CLIPping the Deception: Adapting Vision- Language Models for Universal deepfake detection. In Proceedings of the 2024 International Conference on Multimedia Retrieval, pages 1006–1015, 2024
2024
-
[18]
InfoMax-GAN: Improved Adversarial Image Generation via Information Maximization and Contrastive Learning
Kwot Sin Lee, Ngoc-Trung Tran, and Ngai-Man Cheung. InfoMax-GAN: Improved Adversarial Image Generation via Information Maximization and Contrastive Learning. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pages 3942–3952, 2021
2021
-
[19]
MMD GAN: Towards Deeper Understanding of Moment Matching Network
Chun-Liang Li, Wei-Cheng Chang, Yu Cheng, Yiming Yang, and Barnabás Póczos. MMD GAN: Towards Deeper Understanding of Moment Matching Network. Advances in neural information processing systems, 30, 2017
2017
-
[20]
Detecting Generated Images by Real Images
Bo Liu, Fan Yang, Xiuli Bi, Bin Xiao, Weisheng Li, and Xinbo Gao. Detecting Generated Images by Real Images. In European Conference on Computer Vision, pages 95–110. Springer, 2022
2022
-
[21]
Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection
Huan Liu, Zichang Tan, Chuangchuang Tan, Yunchao Wei, Jingdong Wang, and Yao Zhao. Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 10770–10780, 2024
2024
-
[22]
Pseudo Numerical Methods for Diffusion Models on Manifolds
Luping Liu, Yi Ren, Zhijie Lin, and Zhou Zhao. Pseudo Numerical Methods for Diffusion Models on Manifolds. In International Conference on Learning Representations, 2022
2022
-
[23]
STGAN: A Unified Selective Transfer Network for Arbitrary Image Attribute Editing
Ming Liu, Yukang Ding, Min Xia, Xiao Liu, Errui Ding, Wangmeng Zuo, and Shilei Wen. STGAN: A Unified Selective Transfer Network for Arbitrary Image Attribute Editing. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 3673–3682, 2019
2019
-
[24]
High-Fidelity Image Generation with Fewer Labels
Mario Luˇci´c, Michael Tschannen, Marvin Ritter, Xiaohua Zhai, Olivier Bachem, and Sylvain Gelly. High-Fidelity Image Generation with Fewer Labels. In International conference on machine learning, pages 4183–4192. PMLR, 2019
2019
-
[25]
Generalizing Face Forgery Detection with High-Frequency Features
Yuchen Luo, Yong Zhang, Junchi Yan, and Wei Liu. Generalizing Face Forgery Detection with High-Frequency Features. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16317–16326, 2021
2021
-
[26]
Spectral Normaliza- tion for Generative Adversarial Networks
Takeru Miyato, Toshiki Kataoka, Masanori Koyama, and Yuichi Yoshida. Spectral Normaliza- tion for Generative Adversarial Networks. In International Conference on Learning Represen- tations, 2018
2018
-
[27]
Improved Denoising Fiffusion Probabilistic Models
Alexander Quinn Nichol and Prafulla Dhariwal. Improved Denoising Fiffusion Probabilistic Models. In International conference on machine learning, pages 8162–8171. PMLR, 2021
2021
-
[28]
Glide: Towards photorealistic image genera- tion and editing with text-guided diffusion models
Alexander Quinn Nichol, Prafulla Dhariwal, Aditya Ramesh, Pranav Shyam, Pamela Mishkin, Bob Mcgrew, Ilya Sutskever, and Mark Chen. Glide: Towards photorealistic image genera- tion and editing with text-guided diffusion models. In International Conference on Machine Learning, p...
2022
-
[29]
RelGAN: Relational Generative Adversarial Networks for Text Generation
Weili Nie, Nina Narodytska, and Ankit Patel. RelGAN: Relational Generative Adversarial Networks for Text Generation. In International conference on learning representations, 2018
2018
-
[30]
Towards Universal Fake Image Detectors that Generalize Across Generative Models
Utkarsh Ojha, Yuheng Li, and Yong Jae Lee. Towards Universal Fake Image Detectors that Generalize Across Generative Models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 24480–24489, 2023
2023
-
[31]
Semantic Image Synthe- sis with Spatially-Adaptive Normalization
Taesung Park, Ming-Yu Liu, Ting-Chun Wang, and Jun-Yan Zhu. Semantic Image Synthe- sis with Spatially-Adaptive Normalization. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2337–2346, 2019. 11
2019
-
[32]
Thinking in Frequency: Face Forgery Detection by Mining Frequency-Aware Clues
Yuyang Qian, Guojun Yin, Lu Sheng, Zixuan Chen, and Jing Shao. Thinking in Frequency: Face Forgery Detection by Mining Frequency-Aware Clues. In European conference on computer vision, pages 86–103. Springer, 2020
2020
-
[33]
Zero-Shot Text-to-Image Generation
Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea V oss, Alec Radford, Mark Chen, and Ilya Sutskever. Zero-Shot Text-to-Image Generation. In International conference on machine learning, pages 8821–8831. PMLR, 2021
2021
-
[34]
High- Resolution Image Synthesis with Latent Diffusion Models
Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High- Resolution Image Synthesis with Latent Diffusion Models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022
2022
-
[35]
A mathematical theory of communication
Claude Elwood Shannon. A mathematical theory of communication. The Bell system technical journal, 27(3):379–423, 1948
1948
-
[36]
GANGen-Detection: A Dataset Generated by GANs for Generalizable Deepfake Detection
Chuangchuang Tan, Renshuai Tao, Huan Liu, and Yao Zhao. GANGen-Detection: A Dataset Generated by GANs for Generalizable Deepfake Detection. github.com/chuangchuangtan/ GANGen-Detection, 2024
2024
-
[37]
Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Domain Learning
Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, Ping Liu, and Yunchao Wei. Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Domain Learning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 5052–5060, 2024
2024
-
[38]
Re- thinking the Up-Sampling Operations in CNN-Based Generative Network for Generalizable Deepfake Detection
Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, Ping Liu, and Yunchao Wei. Re- thinking the Up-Sampling Operations in CNN-Based Generative Network for Generalizable Deepfake Detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, p...
2024
-
[39]
Learning on Gradients: Generalized Artifacts Representation for GAN-Generated Images Detection
Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, and Yunchao Wei. Learning on Gradients: Generalized Artifacts Representation for GAN-Generated Images Detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12105–12114, 2023
2023
-
[40]
Visualizing data using t-SNE
Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-SNE. Journal of Machine Learning Research, 9(11), 2008
2008
-
[41]
Cnn- generated images are surprisingly easy to spot
Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, and Alexei A Efros. Cnn- generated images are surprisingly easy to spot... for now. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8695–8704, 2020
2020
-
[42]
DIRE for Diffusion-Generated Image Detection
Zhendong Wang, Jianmin Bao, Wengang Zhou, Weilun Wang, Hezhen Hu, Hong Chen, and Houqiang Li. DIRE for Diffusion-Generated Image Detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 22445–22455, 2023
2023
-
[43]
Combating Misinformation in the Era of Generative AI Models
Danni Xu, Shaojing Fan, and Mohan Kankanhalli. Combating Misinformation in the Era of Generative AI Models. InProceedings of the 31st ACM International Conference on Multimedia, MM ’23, page 9291–9298, 2023
2023
-
[44]
LSUN: Construction of a Large-Scale Image Dataset using Deep Learning with Humans in the Loop
Fisher Yu, Ari Seff, Yinda Zhang, Shuran Song, Thomas Funkhouser, and Jianxiong Xiao. LSUN: Construction of a Large-Scale Image Dataset using Deep Learning with Humans in the Loop. arXiv preprint arXiv:1506.03365, 2015
2015 arXiv
-
[45]
Breaking Semantic Artifacts for Generalized AI-generated Image Detection
Chende Zheng, Chenhao Lin, Zhengyu Zhao, Hang Wang, Xu Guo, Shuai Liu, and Chao Shen. Breaking Semantic Artifacts for Generalized AI-generated Image Detection. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024
2024
-
[46]
Patchcraft: Exploring Texture Patch for Efficient AI-generated Image Detection
Nan Zhong, Yiran Xu, Sheng Li, Zhenxing Qian, and Xinpeng Zhang. Patchcraft: Exploring Texture Patch for Efficient AI-generated Image Detection. arXiv preprint arXiv:2311.12397, pages 1–18, 2024
2024 arXiv
-
[47]
Unpaired Image-to-Image Trans- lation using Cycle-Consistent Adversarial Networks
Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A Efros. Unpaired Image-to-Image Trans- lation using Cycle-Consistent Adversarial Networks. In Proceedings of the IEEE international conference on computer vision, pages 2223–2232, 2017. 12
2017
Reviewed August 16, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.