Skip to content
conftrace_

prefilling phase

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

long-sequence inference (1) decoding phase (1) layer-wise skipping (1) sublayer skipping (1) transformer model (1) computational budget (1) long-context inference (1) inference acceleration (1) sparse attention (1) layer skipping (1)

Papers

AdaSkip: Adaptive Sublayer Skipping for Accelerating Long-Context LLM Inference AAAI 2025 FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference ICLR 2025