Attention May Not Be What You Need

Read full text
Authors: Michael Bonsignore
Cited by: 0

Abstract

A series of ablations asking what’s actually necessary in transformer attention. Each step removed something thought to be essential. Nothing broke. It got better.

Fields of Study & Subjects

Topic Modeling Advanced Neural Network Applications Natural Language Processing Techniques

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
2 February 2026
Language
EN