Gated linear attention Transformers achieve competitive language modeling results with linear-time inference, superior length generalization, and higher training throughput than Mamba.
Hierarchically gated recurrent neural network for sequence modeling
2 Pith papers cite this work, alongside 25 external citations. Polarity classification is still indexing.
2
Pith papers citing it
25
external citations · external index
fields
cs.LG 2representative citing papers
Making memory decay input-dependent via a lightweight MLP improves log-linear attention performance on associative recall, selective copying, and language modeling, especially for long sequences.
citing papers explorer
-
Gated Linear Attention Transformers with Hardware-Efficient Training
Gated linear attention Transformers achieve competitive language modeling results with linear-time inference, superior length generalization, and higher training throughput than Mamba.
-
Adaptive Memory Decay for Log-Linear Attention
Making memory decay input-dependent via a lightweight MLP improves log-linear attention performance on associative recall, selective copying, and language modeling, especially for long sequences.