FlashAttention reduces GPU high-bandwidth memory accesses in self-attention via tiling, delivering exact attention with lower IO complexity, 2-3x wall-clock speedups on models like GPT-2, and the ability to train on sequences up to 64K long.
Movement pruning: Adaptive sparsity by fine-tuning
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
representative citing papers
Widthwise pruning of LVLM language backbones combined with supervised finetuning and hidden-state distillation recovers over 95% performance using just 5% of data across 3B-7B models.
Sparse neural networks achieve better area and energy efficiency when executed on dense matrix multiplication accelerators using a Sparse-on-Dense approach than on dedicated sparse accelerators.
citing papers explorer
-
FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
FlashAttention reduces GPU high-bandwidth memory accesses in self-attention via tiling, delivering exact attention with lower IO complexity, 2-3x wall-clock speedups on models like GPT-2, and the ability to train on sequences up to 64K long.
-
Structural Pruning of Large Vision Language Models: A Comprehensive Study on Pruning Dynamics, Recovery, and Data Efficiency
Widthwise pruning of LVLM language backbones combined with supervised finetuning and hidden-state distillation recovers over 95% performance using just 5% of data across 3B-7B models.
-
Sparse-on-Dense: Area and Energy-Efficient Computing of Sparse Neural Networks on Dense Matrix Multiplication Accelerators
Sparse neural networks achieve better area and energy efficiency when executed on dense matrix multiplication accelerators using a Sparse-on-Dense approach than on dedicated sparse accelerators.