AI RESEARCH
Memory Inception: Latent-Space KV Cache Manipulation for Steering LLMs
arXiv CS.LG
•
ArXi:2605.06225v1 Announce Type: new Steering large language models (LLMs) is usually done by either instruction prompting or activation steering. Prompting often gives strong control, but caches guidance tokens at every layer and can clutter long interactions; activation steering is compact but typically weaker and does not large structured reminders. We