A cluster-masked autoregressive plus multi-task pretraining framework lifts xLSTM vision backbones to 83.4% top-1 accuracy on ImageNet-1K, roughly one point above the ViL-B baseline.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MAL: Cluster-Masked and Multi-Task Pretraining for Enhanced xLSTM Vision Performance
A cluster-masked autoregressive plus multi-task pretraining framework lifts xLSTM vision backbones to 83.4% top-1 accuracy on ImageNet-1K, roughly one point above the ViL-B baseline.