MétaCan
Menu
Back to cohort

Black-Box Watermark Removal Using Diffusion Models and Self-Attention Mechanisms

2025· article· W7125838433 on OpenAlexfundno aff
Qingyuan Zeng, Yunpeng Gong, Chuangliang Zhang, Shu Jiang, Zhenzhong Wang, Min Jiang

Bibliographic record

Venuenot available
Typearticle
Language
FieldComputer Science
TopicAdvanced Steganography and Watermarking Techniques
Canadian institutionsnot available
FundersNatural Sciences and Engineering Research Council of Canada
KeywordsWatermarkDigital watermarkingRobustness (evolution)EmbeddingNoise (video)Image qualityFeature (linguistics)Image (mathematics)

Abstract

fetched live from OpenAlex

Digital watermarking has become a cornerstone for copyright protection in the digital era, yet its robustness against removal attacks remains a critical challenge. Conventional watermark removal techniques often degrade image quality or fail to counter advanced watermarking methods. This paper introduces a novel Diffusion-Based Watermark Removal Attack (DWRA), harnessing the generative power of diffusion models to effectively erase watermarks while preserving image fidelity. Operating in a black-box setting, DWRA requires no prior knowledge of watermark embedding or detection mechanisms. By projecting watermarked images into a deep feature space, applying noise to disrupt watermark features, and employing a self-attention diffusion model for denoising, our approach achieves watermark removal with minimal quality loss. Extensive experiments across diverse watermarking techniques demonstrate that DWRA surpasses existing methods in both removal efficacy and image quality preservation, exposing vulnerabilities in current watermarking systems and emphasizing the need for more resilient designs.

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 machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.001
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.001
Threshold uncertainty score0.003

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0000.000
Science and technology studies0.0000.001
Scholarly communication0.0010.002
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0010.000

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.

Opus teacher head0.016
GPT teacher head0.257
Teacher spread0.241 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicAdvanced Steganography and Watermarking TechniquesFrench-language works237,207