LATIDIA · Ciberseguridad

Marca de agua LLM de co-servicio adaptativo en motores de inferencia modernos

arXiv: 2610.03955v1Tipo de anuncio: nuevo Resumen: La marca de agua del modelo de lenguaje grande (LLM) es importante para la verificación de la propiedad y la protección de la propiedad intelectual. Sin embargo, los enfoques existentes se centran en el desi algorítmico

La noticia

arXiv:2610.03955v1 Announce Type: new Abstract: Large language model (LLM) watermarking is important for ownership verification and intellectual property protection. However, existing approaches focus on algorithmic design while treating LLM inference engines as separate components. This separation often introduces auxiliary models or external tools, increasing latency and memory overhead while limiting the use of modern inference optimizations. As a result, a deployment gap remains: practical watermarking must preserve utility, detectability, and robustness while minimizing serving overhead. To bridge this gap, we propose SWIFT, a framework that co-designs LLM watermarking with modern inference infrastructure. SWIFT (i) integrates text generation and watermark construction within a shared LLM backend to reduce re-computation, improve cache