Showing cs.CRShow all
2 papers · 1 filter
cs.CR2026
SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling
Haotian Xu, Zeyang Zhang, Linbao Li +3
Speculative inference accelerates large language model (LLM) decoding but provides no inherent safety guarantees. Existing safety defenses are largely incompatible with speculative…
cs.CR2025
Context-Aware Hierarchical Learning: A Two-Step Paradigm towards Safer LLMs
Tengyun Ma, Jiaqi Yao, Daojing He +4
Large Language Models (LLMs) have emerged as powerful tools for diverse applications. However, their uniform token processing paradigm introduces critical vulnerabilities in instru…