LATIDIA · Ciberseguridad
AttentionSwitch: Defensa contra el «jailbreaking» para los LLM mediante el aprendizaje conjunto de la selección de cabeceras de atención dispersas y el escalado
arXiv:2611.08752v2 Tipo de anuncio: sustitución cruzada Resumen: Los grandes modelos de lenguaje han logrado un rendimiento notable en una amplia gama de tareas, pero siguen siendo vulnerables a los ataques de «jailbreak» que eluden las medidas de seguridad.
WhatsApp ↗Telegram ↗
La noticia
arXiv:2611.08752v2 Announce Type: replace-cross Abstract: Large language models have achieved remarkable performance across a wide range of tasks, yet they remain susceptible to jailbreak attacks that circumvent safety mechanisms and elicit harmful or unintended outputs. Although safety can be improved through model fine-tuning, such approaches can be computationally expensive and may degrade previously acquired capabilities through catastrophic forgetting. Recent jailbreak defenses have therefore explored targeted interventions on specific model components associated with unsafe behavior. However, existing approaches typically decouple two inherently related decisions: where to intervene and how to modify the selected components. Such sequential optimization may be suboptimal because the optimal intervention location can depend on the intervention applied