Yingfa Chen

@yingfa-chen · 1 works

Builds memory-efficient methods for compressing key-value caches in long-context language models.

Recent posts by Yingfa Chen

• Latent Compaction with Test-Time Training - 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… · Paper · arxiv.org