Anna Cordoba
@anna-cordoba · 1 works
Researches compression methods for long-context language model inference and KV cache efficiency.
Recent posts by Anna Cordoba
- DepthWeave-KV: Token-Adaptive Cross-Layer Residual Factorization for Long-Context KV Cache Compression - Long-context language model inference is increasingly limited by the memory bandwidth and capacity required to store key-value caches, yet existing compression methods often apply uniform budgets… · Paper · arxiv.org