English
Related papers

Related papers: Stochastic Sparse Attention for Memory-Bound Infer…

200 papers

Long contexts improve capabilities of large language models but pose serious hardware challenges: compute and memory footprints grow linearly with sequence length. Particularly, the decoding phase continuously accesses massive KV cache,…

Hardware Architecture · Computer Science 2026-04-29 Wang Fan , Wei Cao , Xi Zha , Kedi Ma , MingQian Sun , Jialin Chen , Fengzhe Zhang , Fan Zhang

Visual AutoRegressive (VAR) modeling has garnered significant attention for its innovative next-scale prediction paradigm. However, mainstream VAR paradigms attend to all tokens across historical scales at each autoregressive step. As the…

Computer Vision and Pattern Recognition · Computer Science 2026-03-31 Zekun Li , Ning Wang , Tongxin Bai , Changwang Mei , Peisong Wang , Shuang Qiu , Jian Cheng

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…

Efficient inference is critical for long-context language models, where attention computation and KV-cache access dominate the cost. Recent work RAT+, introduces a recurrence-augmented attention backbone that enables flexible dilated…

Machine Learning · Computer Science 2026-05-28 Xiuying Wei , Caglar Gulcehre

Vision transformer has achieved impressive performance for many vision tasks. However, it may suffer from high redundancy in capturing local features for shallow layers. Local self-attention or early-stage convolutions are thus utilized,…

Computer Vision and Pattern Recognition · Computer Science 2024-01-26 Huaibo Huang , Xiaoqiang Zhou , Jie Cao , Ran He , Tieniu Tan

Gating mechanisms have been widely utilized, from early models like LSTMs and Highway Networks to recent state space models, linear attention, and also softmax attention. Yet, existing literature rarely examines the specific effects of…

Computation and Language · Computer Science 2025-05-13 Zihan Qiu , Zekun Wang , Bo Zheng , Zeyu Huang , Kaiyue Wen , Songlin Yang , Rui Men , Le Yu , Fei Huang , Suozhi Huang , Dayiheng Liu , Jingren Zhou , Junyang Lin

Attention is the dominant source of latency during long-context LLM inference, an increasingly popular workload with reasoning models and RAG. We propose Kascade, a training-free sparse attention method that leverages known observations…

Machine Learning · Computer Science 2025-12-19 Dhruv Deshmukh , Saurabh Goyal , Nipun Kwatra , Ramachandran Ramjee

We introduce Sparse Forcing, a training-and-inference paradigm for autoregressive video diffusion models that improves long-horizon generation quality while reducing decoding latency. Sparse Forcing is motivated by an empirical observation…

Computer Vision and Pattern Recognition · Computer Science 2026-04-24 Boxun Xu , Yuming Du , Zichang Liu , Siyu Yang , Ziyang Jiang , Siqi Yan , Rajasi Saha , Albert Pumarola , Wenchen Wang , Peng Li

The quadratic complexity of self-attention in Transformer models remains a significant bottleneck for processing long sequences and deploying large language models efficiently. For this approach, there has been significant research into…

Computation and Language · Computer Science 2026-05-26 Spandan Pratyush

Various forms of sparse attention have been explored to mitigate the quadratic computational and memory cost of the attention mechanism in transformers. We study sparse transformers not through a lens of efficiency but rather in terms of…

Machine Learning · Computer Science 2025-06-19 Parikshit Ram , Kenneth L. Clarkson , Tim Klinger , Shashanka Ubaru , Alexander G. Gray

Many advanced Large Language Model (LLM) applications require long-context processing, but the self-attention module becomes a bottleneck during the prefilling stage of inference due to its quadratic time complexity with respect to sequence…

Machine Learning · Computer Science 2025-06-02 Xiaodong Ji , Hailin Zhang , Fangcheng Fu , Bin Cui

Self-attention dominates the computational and memory cost of long-context LLM inference across both prefill and decode phases. To address this challenge, we introduce Sketch&Walk Attention, a training-free sparse attention method that…

Machine Learning · Computer Science 2026-02-10 Hoang Anh Duy Le , Sahil Joshi , Zeyu Yang , Zhaozhuo Xu , Anshumali Shrivastava

Video understanding in multimodal language models remains limited by context length: models often miss key transition frames and struggle to maintain coherence across long time scales. To address this, we adapt Native Sparse Attention (NSA)…

Computer Vision and Pattern Recognition · Computer Science 2026-02-02 Enxin Song , Wenhao Chai , Shusheng Yang , Ethan Armand , Xiaojun Shan , Haiyang Xu , Jianwen Xie , Zhuowen Tu

Language models have demonstrated remarkable capabilities in reasoning tasks through test-time scaling techniques like best-of-N sampling and tree search. However, these approaches often demand substantial computational resources, creating…

Computation and Language · Computer Science 2026-05-22 Woomin Song , Saket Dingliwal , Sai Muralidhar Jayanthi , Bhavana Ganesh , Jinwoo Shin , Aram Galstyan , Sravan Babu Bodapati

Attention is sparse in vision transformers. We observe the final prediction in vision transformers is only based on a subset of most informative tokens, which is sufficient for accurate image recognition. Based on this observation, we…

Computer Vision and Pattern Recognition · Computer Science 2021-10-27 Yongming Rao , Wenliang Zhao , Benlin Liu , Jiwen Lu , Jie Zhou , Cho-Jui Hsieh

Sparse attention improves LLM inference efficiency by selecting a subset of key-value entries, but at the cost of potential accuracy degradation. In particular, omitting critical KV entries can induce substantial errors in model outputs.…

Machine Learning · Computer Science 2026-05-12 Mohsen Dehghankar , Abolfazl Asudeh

Transformers' quadratic complexity with respect to the input sequence length has motivated a body of work on efficient sparse approximations to softmax. An alternative path, used by entmax transformers, consists of having built-in exact…

Computation and Language · Computer Science 2022-04-22 Marcos Treviso , António Góis , Patrick Fernandes , Erick Fonseca , André F. T. Martins

Vision Transformers (ViT) have shown their competitive advantages performance-wise compared to convolutional neural networks (CNNs) though they often come with high computational costs. To this end, previous methods explore different…

Computer Vision and Pattern Recognition · Computer Science 2023-03-27 Cong Wei , Brendan Duke , Ruowei Jiang , Parham Aarabi , Graham W. Taylor , Florian Shkurti

Sparse attention has been proposed as a way to alleviate the quadratic cost of transformers, a central bottleneck in long-context training. A promising line of work is $\alpha$-entmax attention, a differentiable sparse alternative to…

Machine Learning · Computer Science 2026-04-17 Nuno Gonçalves , Hugo Pitorro , Vlad Niculae , Edoardo Ponti , Lei Li , Andre Martins , Marcos Treviso

Advanced autoregressive (AR) video generation models have improved visual fidelity and interactivity, but the quadratic complexity of attention remains a primary bottleneck for efficient deployment. While existing sparse attention solutions…

Computer Vision and Pattern Recognition · Computer Science 2026-05-26 Chengtao Lv , Yumeng Shi , Yushi Huang , Ruihao Gong , Shen Ren , Wenya Wang