An Efficient Frequency Domain Based Attribution and Detection Network
Bibliographic record
Abstract
People nowadays can easily synthesize high fidelity fake images with different types of image content due to the rapid advances of deep learning technologies. Detecting such images and attributing them to their generative models (GMs) is crucial. Existing deep learning methods attempt to identify and classify GM-specific artifacts but often struggle with content-independence and generalizability. In this paper, we observe that while GMs leave unique artifacts in the frequency domain, they are coupled with the image content. Based on this observation, we propose a novel deep learning-based solution that learns input-adaptive masks to highlight GMs’ artifacts and achieve high accuracy on the synthesized image attribution task. In addition, we observed that GMs’ artifacts in the frequency domain remain intact in sub-images of the original image, and they are even retained when the images are distorted. To further improve the accuracy of the proposed solution, we leverage the characteristics of GMs artifacts in sub-images and distorted images to make our network perform more effectively. Our evaluation results show that our proposed solution outperforms other state-of-the-art methods on unseen image types, showing great generalizability.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".