English
Related papers

Related papers: VSPrefill: Vertical-Slash Sparse Attention with Li…

200 papers

The integration of long-context capabilities with visual understanding unlocks unprecedented potential for Vision Language Models (VLMs). However, the quadratic attention complexity during the pre-filling phase remains a significant…

Computer Vision and Pattern Recognition · Computer Science 2025-05-26 Yucheng Li , Huiqiang Jiang , Chengruidong Zhang , Qianhui Wu , Xufang Luo , Surin Ahn , Amir H. Abdi , Dongsheng Li , Jianfeng Gao , Yuqing Yang , Lili Qiu

Long-context LLMs increasingly rely on extended, reusable prefill prompts for agents and domain Q&A, pushing attention and KV-cache to become the dominant decode-time bottlenecks. While sparse attention reduces computation and transfer…

Machine Learning · Computer Science 2026-04-13 Chuxu Song , Zhencan Peng , Jiuqi Wei , Chuanhui Yang

Long-context inference in decoder-only language models is costly because long prompts are processed during Prefill, cached at every layer, and repeatedly attended to during autoregressive Decode. We introduce \emph{Shallow Prefill, dEEp…

Artificial Intelligence · Computer Science 2026-05-08 Jungsuk Oh , Hyeseo Jeon , Hyunjune Ji , Kyongmin Kong , Jay-Yoon Lee

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

The quadratic computational complexity of standard attention mechanisms presents a severe scalability bottleneck for LLMs in long-context scenarios. While hybrid attention mechanisms combining Full Attention (FA) and Sparse Attention (SA)…

Machine Learning · Computer Science 2026-04-10 Quantong Qiu , Zhiyi Hong , Yi Yang , Haitian Wang , Kebin Liu , Qingqing Dang , Juntao Li , Min Zhang

The KV cache in self-attention has emerged as a major bottleneck in long-context and large-batch inference for LLMs. Existing approaches often treat sparsity prediction and compression as separate modules, relying on auxiliary index…

Machine Learning · Computer Science 2026-03-17 Xu Yang , Jiapeng Zhang , Dongyang Zhao , Guo Chen , Zhuo Tang

Long-sequence processing is a critical capability for modern large language models. However, the self-attention mechanism in the standard Transformer architecture faces severe computational and memory bottlenecks when processing long…

Computation and Language · Computer Science 2025-09-30 Weilin Zhao , Zihan Zhou , Zhou Su , Chaojun Xiao , Yuxuan Li , Yanghao Li , Yudi Zhang , Weilun Zhao , Zhen Li , Yuxiang Huang , Ao Sun , Xu Han , Zhiyuan Liu

Large language models have achieved notable success across various domains, yet efficient inference is still limited by the quadratic computation complexity of the attention mechanism. The inference consists of prefilling and decoding…

Computation and Language · Computer Science 2024-09-24 Junlin Lv , Yuan Feng , Xike Xie , Xin Jia , Qirong Peng , Guiming Xie

Long-context agentic workflows have emerged as a defining use case for large language models, making attention efficiency critical for both inference speed and serving cost. Sparse attention addresses this challenge effectively, and…

Computation and Language · Computer Science 2026-03-13 Yushi Bai , Qian Dong , Ting Jiang , Xin Lv , Zhengxiao Du , Aohan Zeng , Jie Tang , Juanzi 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

Video Large Language Models (VLLMs) incur substantial prefilling cost due to the large number of visual tokens. While attention-based token pruning offers a promising acceleration strategy, applying it at shallow decoder layers often causes…

Computer Vision and Pattern Recognition · Computer Science 2026-03-17 Yingjie Xia , Tao Liu , Jinglei Shi , Qingsong Xie , Heng Guo , Jian Yang , Xi Wang

Large Language Models (LLMs) incur quadratic attention complexity with input length, creating a major time bottleneck in the prefilling stage. Existing acceleration methods largely exploit attention score sparsity by estimating blocks with…

Computation and Language · Computer Science 2026-04-22 Zhiyuan He , Yike Zhang , Chengruidong Zhang , Huiqiang Jiang , Yuqing Yang , Lili Qiu

Long-context models are essential for many applications but face inefficiencies in loading large KV caches during decoding. Prior methods enforce fixed token budgets for sparse attention, assuming a set number of tokens can approximate full…

Machine Learning · Computer Science 2025-02-19 Kan Zhu , Tian Tang , Qinyu Xu , Yile Gu , Zhichen Zeng , Rohan Kadekodi , Liangyu Zhao , Ang Li , Arvind Krishnamurthy , Baris Kasikci

The attention mechanism of a transformer has a quadratic complexity, leading to high inference costs and latency for long sequences. However, attention matrices are mostly sparse, which implies that many entries may be omitted from…

Machine Learning · Computer Science 2025-11-25 Jeffrey Willette , Heejun Lee , Sung Ju Hwang

The Transformer architecture, despite its widespread success, struggles with long-context scenarios due to quadratic computation and linear memory growth. While various linear attention variants mitigate these efficiency constraints by…

Machine Learning · Computer Science 2025-10-02 Yuqi Pan , Yongqi An , Zheng Li , Yuhong Chou , Ruijie Zhu , Xiaohui Wang , Mingxuan Wang , Jinqiao Wang , Guoqi Li

Large language models spend most of their inference cost on attention over long contexts, yet empirical behavior suggests that only a small subset of tokens meaningfully contributes to each query. We formalize this phenomenon by modeling…

Artificial Intelligence · Computer Science 2026-02-17 Vashista Nobaub

State-of-the-art sparse attention methods for reducing decoding latency fall into two main categories: approximate top-$k$ (and its extension, top-$p$) and recently introduced sampling-based estimation. However, these approaches are…

Block-wise sparse attention offers significant efficiency gains for long-context modeling, yet existing methods often suffer from low selection fidelity and cumulative contextual loss by completely discarding unselected blocks. To address…

Computation and Language · Computer Science 2026-02-02 Bailin Wang , Dan Friedman , Tao Lei , Chong Wang

Large Language Models (LLMs) are increasingly deployed in long-context tasks such as reasoning, code generation, and multi-turn dialogue. However, inference over extended contexts is bottlenecked by the Key-Value (KV) cache, whose memory…

Computation and Language · Computer Science 2026-05-21 Seonghwan Choi , Beomseok Kang , Dongwon Jo , Jae-Joon Kim

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