Paper recorded by Signals 4 on 2026-10-14 in cs.LG. Abstract reproduced from arXiv; link to the original below.
Published 2026-10-14 on arXiv · recorded by Signals 4 on 2026-10-15
Category: cs.LG · 机器学习 · first seen 2026-10-15
Long contexts enlarge key--value (KV) caches, increasing memory use and attention costs during generation. Compressing these caches must preserve information for unseen questions. We introduce **Latent Compaction**, which learns a compact document-specific cache through test-time training. We initialize cache entries using reconstruction importance, teacher attention, and key-space diversity. Trai