English
Related papers

Related papers: FlashAttention-2: Faster Attention with Better Par…

200 papers

Large-scale diffusion models such as FLUX (12B parameters) and Stable Diffusion 3 (8B parameters) require multi-GPU parallelism for efficient inference. Unified Sequence Parallelism (USP), which combines Ulysses and Ring attention…

Computer Vision and Pattern Recognition · Computer Science 2026-02-12 Guandong Li

Models such as VGGT and $\pi^3$ have shown strong multi-view 3D performance, but their heavy reliance on global self-attention results in high computational cost. Existing sparse-attention variants offer partial speedups, yet lack a…

Computer Vision and Pattern Recognition · Computer Science 2026-03-11 Xianbing Sun , Zhikai Zhu , Zhengyu Lou , Bo Yang , Jinyang Tang , Liqing Zhang , He Wang , Jianfu Zhang

Diffusion Transformers, particularly for video generation, achieve remarkable quality but suffer from quadratic attention complexity, leading to prohibitive latency. Existing acceleration methods face a fundamental trade-off: dynamically…

Computer Vision and Pattern Recognition · Computer Science 2025-11-17 Dor Shmilovich , Tony Wu , Aviad Dahan , Yuval Domb

Transformers have achieved success in both language and vision domains. However, it is prohibitively expensive to scale them to long sequences such as long documents or high-resolution images, because self-attention mechanism has quadratic…

Computer Vision and Pattern Recognition · Computer Science 2021-12-08 Chen Zhu , Wei Ping , Chaowei Xiao , Mohammad Shoeybi , Tom Goldstein , Anima Anandkumar , Bryan Catanzaro

Nowadays, Large Language Models (LLMs) have been trained using extended context lengths to foster more creative applications. However, long context training poses great challenges considering the constraint of GPU memory. It not only leads…

Machine Learning · Computer Science 2025-01-16 Pinxue Zhao , Hailin Zhang , Fangcheng Fu , Xiaonan Nie , Qibin Liu , Fang Yang , Yuanbo Peng , Dian Jiao , Shuaipeng Li , Jinbao Xue , Yangyu Tao , Bin Cui

To design fast neural networks, many works have been focusing on reducing the number of floating-point operations (FLOPs). We observe that such reduction in FLOPs, however, does not necessarily lead to a similar level of reduction in…

Computer Vision and Pattern Recognition · Computer Science 2023-05-23 Jierun Chen , Shiu-hong Kao , Hao He , Weipeng Zhuo , Song Wen , Chul-Ho Lee , S. -H. Gary Chan

Transformers have emerged as the cornerstone of state-of-the-art natural language processing models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands posed by the self-attention…

Computation and Language · Computer Science 2023-08-30 Hao Liu , Pieter Abbeel

Accommodating long sequences efficiently in autoregressive Transformers, especially within an extended context window, poses significant challenges due to the quadratic computational complexity and substantial KV memory requirements…

Computation and Language · Computer Science 2024-06-25 Chao Lou , Zixia Jia , Zilong Zheng , Kewei Tu

Attention is a key part of the transformer architecture. It is a sequence-to-sequence mapping that transforms each sequence element into a weighted sum of values. The weights are typically obtained as the softmax of dot products between…

Transformers are mostly relying on softmax attention, which introduces quadratic complexity with respect to sequence length and remains a major bottleneck for efficient inference. Prior work on linear or hybrid attention typically replaces…

In both Computer Vision and the wider Deep Learning field, the Transformer architecture is well-established as state-of-the-art for many applications. For Multitask Learning, however, where there may be many more queries necessary compared…

Computer Vision and Pattern Recognition · Computer Science 2025-08-07 Christian Bohn , Thomas Kurbiel , Klaus Friedrichs , Hasan Tercan , Tobias Meisen

Multi-head attention layers, as used in the Transformer neural sequence model, are a powerful alternative to RNNs for moving information across and between sequences. While training these layers is generally fast and simple, due to…

Neural and Evolutionary Computing · Computer Science 2019-11-07 Noam Shazeer

While long-context inference is crucial for advancing large language model (LLM) applications, its prefill speed remains a significant bottleneck. Current approaches, including sequence parallelism strategies and compute reduction through…

Machine Learning · Computer Science 2025-05-27 Yuxiang Huang , Mingye Li , Xu Han , Chaojun Xiao , Weilin Zhao , Sun Ao , Hao Zhou , Jie Zhou , Zhiyuan Liu , Maosong Sun

As the demand for processing extended textual data grows, the ability to handle long-range dependencies and maintain computational efficiency is more critical than ever. One of the key issues for long-sequence modeling using attention-based…

Computation and Language · Computer Science 2025-05-26 Aosong Feng , Rex Ying , Leandros Tassiulas

Speech separation always faces the challenge of handling prolonged time sequences. Past methods try to reduce sequence lengths and use the Transformer to capture global information. However, due to the quadratic time complexity of the…

Audio and Speech Processing · Electrical Eng. & Systems 2025-08-28 Haoxu Wang , Yiheng Jiang , Gang Qiao , Pengteng Shi , Biao Tian

The quadratic complexity and weak length extrapolation of Transformers limits their ability to scale to long sequences, and while sub-quadratic solutions like linear attention and state space models exist, they empirically underperform…

Machine Learning · Computer Science 2024-04-17 Xuezhe Ma , Xiaomeng Yang , Wenhan Xiong , Beidi Chen , Lili Yu , Hao Zhang , Jonathan May , Luke Zettlemoyer , Omer Levy , Chunting Zhou

We present a novel bi-directional Transformer architecture (BiXT) which scales linearly with input size in terms of computational cost and memory consumption, but does not suffer the drop in performance or limitation to only one input…

Computer Vision and Pattern Recognition · Computer Science 2024-11-01 Markus Hiller , Krista A. Ehinger , Tom Drummond

We argue that neither transformers nor sub-quadratic architectures are well suited to training at long sequence lengths: the cost of processing the context is too expensive in the former, too inexpensive in the latter. Approaches such as…

Machine Learning · Computer Science 2025-07-08 Carles Gelada , Jacob Buckman , Sean Zhang , Txus Bach

The quadratic complexity of standard attention mechanisms poses a significant scalability bottleneck for large language models (LLMs) in long-context scenarios. While hybrid attention strategies that combine sparse and full attention within…

Computation and Language · Computer Science 2026-01-29 Zecheng Tang , Quantong Qiu , Yi Yang , Zhiyi Hong , Haiya Xiang , Kebin Liu , Qingqing Dang , Juntao Li , Min Zhang

Looped Transformers (LT) have emerged as a powerful architecture by iterating their layers multiple times before decoding the final token. However, pairing them with full attention retains quadratic complexity, making them computationally…

Machine Learning · Computer Science 2026-05-26 Chunyuan Deng , Yizhe Zhang , Rui-Jie Zhu , Yuanyuan Xu , Jiarui Liu , T. S. Eugene Ng , Hanjie Chen