# paraphrasing
標記為「paraphrasing」的 4 篇文章
Semantic Injection Attacks
Meaning-preserving adversarial attacks that evade syntactic detection by encoding malicious intent in semantically equivalent but structurally different formulations.
semantic-injectionevasionparaphrasingmeaning-preservingdetection-bypass
Watermark Removal Techniques
Techniques for removing AI watermarks: paraphrasing attacks, token substitution, embedding space perturbation, and implications for model provenance and accountability.
watermarkremovalparaphrasingprovenanceaccountabilitydetection-evasion
Semantic Injection 攻擊s
Meaning-preserving adversarial attacks that evade syntactic detection by encoding malicious intent in semantically equivalent but structurally different formulations.
semantic-injectionevasionparaphrasingmeaning-preservingdetection-bypass
浮水印移除技術
移除 AI 浮水印的技術：換句話攻擊、token 替換、embedding 空間擾動，及其對模型來源與可究責性的意涵。
watermarkremovalparaphrasingprovenanceaccountabilitydetection-evasion