相关论文: SPECTRE: An FFT-Based Efficient Drop-In Replacemen…

Large language models have an exceptional capability to incorporate new information in a contextual manner. However, the full potential of such an approach is often restrained due to a limitation in the effective context length. One…

We introduce the Block Transformer which adopts hierarchical global-to-local modeling to autoregressive transformers to mitigate the inference bottlenecks associated with self-attention. Self-attention requires the key-value (KV) cache of…

We present ASSET, a neural architecture for automatically modifying an input high-resolution image according to a user's edits on its semantic segmentation map. Our architecture is based on a transformer with a novel attention mechanism.…

As the foundation of large language models (LLMs), self-attention module faces the challenge of quadratic time and memory complexity with respect to sequence length. FlashAttention accelerates attention computation and reduces its memory…

Transformers have demonstrated remarkable success across vision, language, and video. Yet, increasing task complexity has led to larger models and more tokens, raising the quadratic cost of self-attention and the overhead of GPU memory…

Efficient attention mechanisms enable long-context transformers but often miss globally important tokens, degrading modeling quality. We introduce a pre-scoring framework that assigns a query-independent global importance prior to keys…

As large language models (LLMs) show impressive performance on complex tasks, they still struggle with longer contextual understanding and high computational costs. To balance efficiency and quality, we introduce LLMSteer, a…

To address the high resolution of image pixels, the Swin Transformer introduces window attention. This mechanism divides an image into non-overlapping windows and restricts attention computation to within each window, significantly…

The Transformer architecture has shown significant success in many language processing and visual tasks. However, the method faces challenges in efficiently scaling to long sequences because the self-attention computation is quadratic with…

Large Language Models (LLMs) demonstrate substantial potential across a diverse array of domains via request serving. However, as trends continue to push for expanding context sizes, the autoregressive nature of LLMs results in highly…

We present the first unified study of the efficiency of self-attention-based Transformer variants spanning text, speech and vision. We identify input length thresholds (tipping points) at which efficient Transformer variants become more…

Hyperspectral image super-resolution has attained widespread prominence to enhance the spatial resolution of hyperspectral images. However, convolution-based methods have encountered challenges in harnessing the global spatial-spectral…

Long-context video understanding and generation pose a significant computational challenge for Transformer-based video models due to the quadratic complexity of self-attention. While existing sparse attention methods employ coarse-grained…

Transformers, driven by attention mechanisms, form the foundation of large language models (LLMs). As these models scale up, efficient GPU attention kernels become essential for high-throughput and low-latency inference. Diverse LLM…

Attention is a core component of transformer architecture, whether encoder-only, decoder-only, or encoder-decoder model. However, the standard softmax attention often produces noisy probability distribution, which can impair effective…

Large Language Models (LLMs) face efficiency bottlenecks due to the quadratic complexity of the attention mechanism when processing long contexts. Sparse attention methods offer a promising solution, but existing approaches often suffer…

FlashAttention (Dao, 2023) effectively reduces the quadratic peak memory usage to linear in training transformer-based large language models (LLMs) on a single GPU. In this paper, we introduce DISTFLASHATTN, a distributed memory-efficient…

Squeeze-and-Excitation (SE) block presents a channel attention mechanism for modeling global context via explicitly capturing dependencies across channels. However, we are still far from understanding how the SE block works. In this work,…

Transformers and large language models (LLMs) have revolutionized machine learning, with attention mechanisms at the core of their success. As the landscape of attention variants expands, so too do the challenges of optimizing their…

Efficient inference on GPUs using large language models remains challenging due to memory bandwidth limitations, particularly during data transfers between High Bandwidth Memory (HBM) and SRAM in attention computations. Approximate…