all AI news
Softmax Attention with Constant Cost per Token
April 10, 2024, 4:41 a.m. | Franz A. Heinsen
cs.LG updates on arXiv.org arxiv.org
Abstract: We propose a simple modification to the conventional attention mechanism applied by Transformers: Instead of quantifying pairwise query-key similarity with scaled dot-products, we quantify it with the logarithms of scaled dot-products of exponentials. Attention becomes expressible as a composition of log-sums of exponentials that is linearizable, with a latent space of constant size, enabling sequential application with constant time and space complexity per token. We implement our modification, verify that it works in practice, and …
abstract arxiv attention cost cs.cl cs.lg key per products query simple softmax token transformers type
More from arxiv.org / cs.LG updates on arXiv.org
Jobs in AI, ML, Big Data
AI Research Scientist
@ Vara | Berlin, Germany and Remote
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
Senior Data Scientist
@ ITE Management | New York City, United States