TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization

Open in new window