Breaking Semantic-Aware Watermarks via LLM-Guided Coherence-Preserving Semantic Injection
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908852081917952 |
|---|---|
| author | Gao, Zheng Li, Xiaoyu Bao, Zhicheng Feng, Xiaoyan Jiang, Jiaojiao |
| author_facet | Gao, Zheng Li, Xiaoyu Bao, Zhicheng Feng, Xiaoyan Jiang, Jiaojiao |
| contents | Generative images have proliferated on Web platforms in social media and online copyright distribution scenarios, and semantic watermarking has increasingly been integrated into diffusion models to support reliable provenance tracking and forgery prevention for web content. Traditional noise-layer-based watermarking, however, remains vulnerable to inversion attacks that can recover embedded signals. To mitigate this, recent content-aware semantic watermarking schemes bind watermark signals to high-level image semantics, constraining local edits that would otherwise disrupt global coherence. Yet, large language models (LLMs) possess structured reasoning capabilities that enable targeted exploration of semantic spaces, allowing locally fine-grained but globally coherent semantic alterations that invalidate such bindings. To expose this overlooked vulnerability, we introduce a Coherence-Preserving Semantic Injection (CSI) attack that leverages LLM-guided semantic manipulation under embedding-space similarity constraints. This alignment enforces visual-semantic consistency while selectively perturbing watermark-relevant semantics, ultimately inducing detector misclassification. Extensive empirical results show that CSI consistently outperforms prevailing attack baselines against content-aware semantic watermarking, revealing a fundamental security weakness of current semantic watermark designs when confronted with LLM-driven semantic perturbations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_21593 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Breaking Semantic-Aware Watermarks via LLM-Guided Coherence-Preserving Semantic Injection Gao, Zheng Li, Xiaoyu Bao, Zhicheng Feng, Xiaoyan Jiang, Jiaojiao Machine Learning Cryptography and Security Computer Vision and Pattern Recognition Generative images have proliferated on Web platforms in social media and online copyright distribution scenarios, and semantic watermarking has increasingly been integrated into diffusion models to support reliable provenance tracking and forgery prevention for web content. Traditional noise-layer-based watermarking, however, remains vulnerable to inversion attacks that can recover embedded signals. To mitigate this, recent content-aware semantic watermarking schemes bind watermark signals to high-level image semantics, constraining local edits that would otherwise disrupt global coherence. Yet, large language models (LLMs) possess structured reasoning capabilities that enable targeted exploration of semantic spaces, allowing locally fine-grained but globally coherent semantic alterations that invalidate such bindings. To expose this overlooked vulnerability, we introduce a Coherence-Preserving Semantic Injection (CSI) attack that leverages LLM-guided semantic manipulation under embedding-space similarity constraints. This alignment enforces visual-semantic consistency while selectively perturbing watermark-relevant semantics, ultimately inducing detector misclassification. Extensive empirical results show that CSI consistently outperforms prevailing attack baselines against content-aware semantic watermarking, revealing a fundamental security weakness of current semantic watermark designs when confronted with LLM-driven semantic perturbations. |
| title | Breaking Semantic-Aware Watermarks via LLM-Guided Coherence-Preserving Semantic Injection |
| topic | Machine Learning Cryptography and Security Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.21593 |