A framework to identify and convert foldable layer normalizations to RMSNorm for exact equivalence and faster inference in deep neural networks.
How to train your vit? data, augmentation, and regularization in vision transformers
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
π₀ is a vision-language-action flow model trained on diverse multi-platform robot data that supports zero-shot task performance, language instruction following, and efficient fine-tuning for dexterous tasks.
MetaCLIP curates balanced 400M-pair subsets from CommonCrawl that outperform CLIP data, reaching 70.8% zero-shot ImageNet accuracy on ViT-B versus CLIP's 68.3%.
SigLIP replaces softmax-based contrastive loss with a simple pairwise sigmoid loss for vision-language pre-training, decoupling batch size from normalization and reaching strong zero-shot performance with limited compute.
DAP improves ViT attribution maps by injecting decision-relevant gradients into attention propagation, producing more class-sensitive and faithful explanations than standard attention rollout.
A comprehensive survey of PEFT algorithms for large models, covering their performance, overhead, applications, and real-world system implementations.
citing papers explorer
-
Enjoy Your Layer Normalization with the Computational Efficiency of RMSNorm
A framework to identify and convert foldable layer normalizations to RMSNorm for exact equivalence and faster inference in deep neural networks.
-
$\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
π₀ is a vision-language-action flow model trained on diverse multi-platform robot data that supports zero-shot task performance, language instruction following, and efficient fine-tuning for dexterous tasks.
-
Demystifying CLIP Data
MetaCLIP curates balanced 400M-pair subsets from CommonCrawl that outperform CLIP data, reaching 70.8% zero-shot ImageNet accuracy on ViT-B versus CLIP's 68.3%.
-
Sigmoid Loss for Language Image Pre-Training
SigLIP replaces softmax-based contrastive loss with a simple pairwise sigmoid loss for vision-language pre-training, decoupling batch size from normalization and reaching strong zero-shot performance with limited compute.
-
Decision-Aware Attention Propagation for Vision Transformer Explainability
DAP improves ViT attribution maps by injecting decision-relevant gradients into attention propagation, producing more class-sensitive and faithful explanations than standard attention rollout.
-
Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey
A comprehensive survey of PEFT algorithms for large models, covering their performance, overhead, applications, and real-world system implementations.