conftrace_

global-to-local modeling

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

batch inference (1) hierarchical attention (1) key-value cache (1) kv cache (1) inference throughput (1) inference acceleration (1) inference efficiency (1) language model (1) autoregressive generation (1) hierarchical modeling (1)

Papers

Block Transformer: Global-to-Local Language Modeling for Fast Inference NeurIPS 2024