Guilin Zhang

Lazy-loading agent skills on demand doesn't just slash context window costs—it actively sharpens tool retrieval quality, all while incurring only a negligible hit to latency.

Turns out, your hand-tuned autoscaler probably beats that fancy RL agent you're trying to deploy for adaptive resource control.

Forget opaque LLM-driven memory policies: A-MAC offers transparent, efficient control over LLM agent memory by factoring in utility, confidence, novelty, recency, and content type.