conftrace_

attention approximation

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

softmax kernel (1) large language model (1) sequence length (1) decoding efficiency (1) quadratic complexity (1) kv cache (1) sampling method (1) random feature (1) approximation algorithm (1) nyström method (1)

Papers

MagicPIG: LSH Sampling for Efficient LLM Generation ICLR 2025 Dense-Exponential Random Features: Sharp Positive Estimators of the Gaussian Kernel NeurIPS 2023 Nyströmformer: A Nyström-based Algorithm for Approximating Self-Attention AAAI 2021