English
Related papers

Related papers: FlashEVA: Accelerating LLM inference via Efficient…

200 papers

Transformers have outperformed recurrent neural networks (RNNs) in natural language generation. But this comes with a significant computational cost, as the attention mechanism's complexity scales quadratically with sequence length.…

Computation and Language · Computer Science 2021-09-21 Jungo Kasai , Hao Peng , Yizhe Zhang , Dani Yogatama , Gabriel Ilharco , Nikolaos Pappas , Yi Mao , Weizhu Chen , Noah A. Smith

The vanilla self-attention mechanism in Transformers can be viewed as a two-layer fast-weight MLP, whose weights are dynamically induced by inputs and whose hidden dimension is equal to the sequence length $N$. As the context extends, the…

Machine Learning · Computer Science 2026-05-12 Qishuai Wen , Zhiyuan Huang , Xianghan Meng , Wei He , Chun-Guang Li

Large Language Models (LLMs) are widely used across various domains, processing millions of daily requests. This surge in demand poses significant challenges in optimizing throughput and latency while keeping costs manageable. The Key-Value…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-07-23 Jiale Xu , Rui Zhang , Cong Guo , Weiming Hu , Zihan Liu , Feiyang Wu , Yu Feng , Shixuan Sun , Changxu Shao , Yuhong Guo , Junping Zhao , Ke Zhang , Minyi Guo , Jingwen Leng

The growing demand for long-context inference capabilities in Large Language Models (LLMs) has intensified the computational and memory bottlenecks inherent to the self-attention mechanism. To address this challenge, we introduce BLASST, a…

Currently, one main research line in designing a more efficient vision transformer is reducing the computational cost of self attention modules by adopting sparse attention or using local attention windows. In contrast, we propose a…

Computer Vision and Pattern Recognition · Computer Science 2023-03-21 Haokui Zhang , Wenze Hu , Xiaoyu Wang

Diffusion large language models (dLLMs) are emerging as a promising alternative to autoregressive models (ARMs) due to their ability to capture bidirectional context and the potential for parallel generation. Despite the advantages, dLLM…

Machine Learning · Computer Science 2026-03-12 Zijian Zhu , Fei Ren , Zhanhong Tan , Kaisheng Ma

We show that Transformer encoder architectures can be sped up, with limited accuracy costs, by replacing the self-attention sublayers with simple linear transformations that "mix" input tokens. These linear mixers, along with standard…

Computation and Language · Computer Science 2022-05-30 James Lee-Thorp , Joshua Ainslie , Ilya Eckstein , Santiago Ontanon

Transformer has become ubiquitous in natural language processing (e.g., machine translation, question answering); however, it requires enormous amount of computations to achieve high performance, which makes it not suitable for mobile…

Computation and Language · Computer Science 2020-04-27 Zhanghao Wu , Zhijian Liu , Ji Lin , Yujun Lin , Song Han

Attention, as a core layer of the ubiquitous Transformer architecture, is the bottleneck for large language models and long-context applications. While FlashAttention-3 optimized attention for Hopper GPUs through asynchronous execution and…

Computation and Language · Computer Science 2026-03-06 Ted Zadouri , Markus Hoehnerbach , Jay Shah , Timmy Liu , Vijay Thakkar , Tri Dao

The size and compute characteristics of modern large language models have led to an increased interest in developing specialized kernels tailored for particular training and inference workloads. Existing kernels primarily optimize for…

Machine Learning · Computer Science 2025-12-05 Aniruddha Nrusimha , William Brandon , Mayank Mishra , Yikang Shen , Rameswar Panda , Jonathan Ragan-Kelley , Yoon Kim

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…

Computation and Language · Computer Science 2025-11-11 Dhananjay Ram , Wei Xia , Stefano Soatto

Fine-tuning of self-supervised models is a powerful transfer learning method in a variety of fields, including speech processing, since it can utilize generic feature representations obtained from large amounts of unlabeled data.…

Multimedia · Computer Science 2022-12-07 Shinta Otake , Rei Kawakami , Nakamasa Inoue

Inference with Transformer-based Large Language Models (LLMs) on long sequences is both costly and slow due to the quadratic complexity of the self-attention mechanism. We introduce Star Attention, a two-phase block-sparse approximation…

Computation and Language · Computer Science 2025-06-02 Shantanu Acharya , Fei Jia , Boris Ginsburg

Large Language Models struggle with memory demands from the growing Key-Value (KV) cache as context lengths increase. Existing compression methods homogenize head dimensions or rely on attention-guided token pruning, often sacrificing…

Computation and Language · Computer Science 2025-06-16 Xiaoran Liu , Siyang He , Qiqi Wang , Ruixiao Li , Yuerong Song , Zhigeng Liu , Linlin Li , Qun Liu , Zengfeng Huang , Qipeng Guo , Ziwei He , Xipeng Qiu

Transformers have been matching deep convolutional networks for vision architectures in recent works. Most work is focused on getting the best results on large-scale benchmarks, and scaling laws seem to be the most successful strategy:…

Computer Vision and Pattern Recognition · Computer Science 2023-03-23 Corentin Dancette , Matthieu Cord

Reducing the key-value (KV) cache burden in Large Language Models (LLMs) significantly accelerates inference. Dynamically selecting critical KV caches during decoding helps maintain performance. Existing methods use random linear hashing to…

Computation and Language · Computer Science 2025-10-10 Wenhao Li , Yuxin Zhang , Gen Luo , Haiyuan Wan , Ziyang Gong , Fei Chao , Rongrong Ji

As Large Language Models (LLMs) scale to longer context windows, the computational cost of attention mechanisms, which traditionally grows quadratically with input length, presents a critical challenge for real-time and memory-constrained…

Computation and Language · Computer Science 2024-12-10 James Vo

Autoregressive Transformers rely on Key-Value (KV) caching to accelerate inference. However, the linear growth of the KV cache with context length leads to excessive memory consumption and bandwidth constraints. This bottleneck is…

Computation and Language · Computer Science 2025-06-10 Ravi Ghadia , Avinash Kumar , Gaurav Jain , Prashant Nair , Poulami Das

Over the past 7 years, attention has become one of the most important primitives in deep learning. The primary approach to optimize attention is FlashAttention, which fuses the operation together, drastically improving both the runtime and…

Machine Learning · Computer Science 2024-12-10 Juechu Dong , Boyuan Feng , Driss Guessous , Yanbo Liang , Horace He

Accurate speech emotion recognition is essential for developing human-facing systems. Recent advancements have included finetuning large, pretrained transformer models like Wav2Vec 2.0. However, the finetuning process requires substantial…

Sound · Computer Science 2025-03-07 Aneesha Sampath , James Tavernor , Emily Mower Provost