EPISODE · Nov 24, 2025 · 37 MIN
Performers: Linear Transformers with Orthogonal Random Features
from The Gist Talk · host kw
The provided text introduces Performers, a novel class of Transformer architectures designed to overcome the quadratic time and space complexity limitations of traditional Transformers, which are often prohibitive for long sequences. Performers achieve linear complexity through a mechanism called Fast Attention Via positive Orthogonal Random features (FAVOR+). This approach offers a provably accurate estimation of the standard softmax full-rank attention without requiring priors like sparsity. The paper substantiates its claims with strong theoretical guarantees concerning estimation accuracy and variance reduction, particularly highlighting the necessity of positive random features over unstable trigonometric features. Experimental results confirm that Performers are efficient and effective across various large-scale tasks, including text and protein sequence modeling, often matching or surpassing the performance of other efficient attention methods
What this episode covers
The provided text introduces Performers, a novel class of Transformer architectures designed to overcome the quadratic time and space complexity limitations of traditional Transformers, which are often prohibitive for long sequences. Performers achieve linear complexity through a mechanism called Fast Attention Via positive Orthogonal Random features (FAVOR+). This approach offers a provably accurate estimation of the standard softmax full-rank attention without requiring priors like sparsity. The paper substantiates its claims with strong theoretical guarantees concerning estimation accuracy and variance reduction, particularly highlighting the necessity of positive random features over unstable trigonometric features. Experimental results confirm that Performers are efficient and effective across various large-scale tasks, including text and protein sequence modeling, often matching or surpassing the performance of other efficient attention methods
NOW PLAYING
Performers: Linear Transformers with Orthogonal Random Features
No transcript for this episode yet
Similar Episodes
Mar 26, 2026 ·1m
Mar 19, 2026 ·34m
Feb 18, 2026 ·11m
Feb 11, 2026 ·45m