betaFeedback

WorksComputer Science

LazyFlash: Forward-Dense, Backward-Sparse Attention for Accelerated Long-Context Fine-Tuning

Abstract

As large language models (LLMs) process increasingly long sequences, attention becomes a major computational bottleneck. Sparse attention mostly targets inference, yet many methods still train densely or cannot support sparse training, and their added complexity makes sparsity hard to turn into wall-clock speedups, leaving dense attention widely used in practice. This paper introduces LazyFlash, a forward-dense, backward-sparse attention method built on the observation that attention training is dominated by the backward pass. LazyFlash keeps the forward pass exactly dense, so the trained model remains a standard dense model that existing systems such as vLLM can serve without any change. For the backward pass, LazyFlash proposes a gradient-aware block sparsification design that ranks query blocks by their attention-output gradient norms and skips those contributing negligibly to parameter updates, turning the dense backward into a block-sparse one. A gradient-mass budget adapts the amount of skipped computation to each head and training step, while importance sampling preserves the dense gradient in expectation. These designs are implemented inside the tiled FlashAttention backward kernel, where skipped blocks are never loaded or computed, translating gradient sparsity into real wall-clock acceleration. Experiments on long-context text and speech LLMs show that LazyFlash maintains downstream performance while achieving up to 5.7x attention-backward speedup over FlashAttention and 1.7x end-to-end training speedup.

Paper details

vlnc:W_damztn8w
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 10

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

generative AI tools

View statement
Looking for related papers…