English
Related papers

Related papers: HySparse: A Hybrid Sparse Attention Architecture w…

200 papers

The quadratic complexity of self-attention during the prefill phase impedes long-context inference in large language models. Existing sparse attention methods face a trade-off among context adaptivity, sampling overhead, and fine-tuning…

Machine Learning · Computer Science 2026-03-06 Chen Guanzhong

Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, compresses key-value states into a low-rank latent vector, caching only this vector to reduce memory. In tensor parallelism (TP), however, attention heads are computed across…

Machine Learning · Computer Science 2025-08-26 Xiaojuan Tang , Fanxu Meng , Pingzhi Tang , Yuxuan Wang , Di Yin , Xing Sun , Muhan Zhang

The efficiency of large vision-language models (LVLMs) is constrained by the computational bottleneck of the attention mechanism during the prefill phase and the memory bottleneck of fetching the key-value (KV) cache in the decoding phase,…

Computer Vision and Pattern Recognition · Computer Science 2024-12-19 Yefei He , Feng Chen , Jing Liu , Wenqi Shao , Hong Zhou , Kaipeng Zhang , Bohan Zhuang

Decoding throughput improvements from larger inference batches are limited by GPU memory, which is largely consumed by the key-value (KV) cache. Prior training-free KV cache offloading alleviates this by keeping redundant context on the CPU…

Computation and Language · Computer Science 2026-01-30 Yuxiang Huang , Pengjie Wang , Jicheng Han , Weilin Zhao , Zhou Su , Ao Sun , Hongya Lyu , Hengyu Zhao , Yudong Wang , Chaojun Xiao , Xu Han , Zhiyuan Liu

Mixture of Experts (MoE) has become a mainstream architecture for building Large Language Models (LLMs) by reducing per-token computation while enabling model scaling. It can be viewed as partitioning a large Feed-Forward Network (FFN) at…

Machine Learning · Computer Science 2025-08-27 Weilin Cai , Le Qin , Shwai He , Junwei Cui , Ang Li , Jiayi Huang

Multimodal Large Language Models (MLLMs) have advanced unified reasoning over text, images, and videos, but their inference is hindered by the rapid growth of key-value (KV) caches. Each visual input expands into thousands of tokens,…

Artificial Intelligence · Computer Science 2026-04-08 Bowen Zeng , Feiyang Ren , Jun Zhang , Xiaoling Gu , Ke Chen , Lidan Shou , Huan Li

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

Cache fusion accelerates generation process of LLMs equipped with RAG through KV caching and selective token recomputation, thereby reducing computational costs and improving efficiency. However, existing methods primarily rely on local…

Databases · Computer Science 2026-04-13 Jianxin Yan , Zeheng Qian , Wangze Ni , Zhitao Shen , Zhiping Wang , Haoyang Li , Jia Zhu , Lei Chen , Kui Ren

Modern large language models increasingly require long contexts for reasoning and multi-document tasks, but attention's quadratic complexity creates a severe computational bottleneck. We present Block-Sparse FlashAttention (BSFA), a drop-in…

Machine Learning · Computer Science 2025-12-09 Daniel Ohayon , Itay Lamprecht , Itay Hubara , Israel Cohen , Daniel Soudry , Noam Elata

Trained on 680,000 hours of massive speech data, Whisper is a multitasking, multilingual speech foundation model demonstrating superior performance in automatic speech recognition, translation, and language identification. However, its…

Sound · Computer Science 2024-07-16 Li Zhang , Ning Jiang , Qing Wang , Yue Li , Quan Lu , Lei Xie

Sparse Matrix-Vector multiplication (SpMV) is an essential computational kernel in many application scenarios. Tens of sparse matrix formats and implementations have been proposed to compress the memory storage and speed up SpMV…

Distributed, Parallel, and Cluster Computing · Computer Science 2022-12-22 Zhen Du , Jiajia Li , Yinshan Wang , Xueqi Li , Guangming Tan , Ninghui Sun

Transformers provide a class of expressive architectures that are extremely effective for sequence modeling. However, the key limitation of transformers is their quadratic memory and time complexity $\mathcal{O}(L^2)$ with respect to the…

Machine Learning · Computer Science 2021-10-29 Hongyu Ren , Hanjun Dai , Zihang Dai , Mengjiao Yang , Jure Leskovec , Dale Schuurmans , Bo Dai

Large language models have revolutionized data processing in numerous domains, with their ability to handle extended context reasoning receiving notable recognition. To speed up inference, maintaining a key-value (KV) cache memory is…

Computation and Language · Computer Science 2024-10-22 Zhen Yang , J. N. Han , Kan Wu , Ruobing Xie , An Wang , Xingwu Sun , Zhanhui Kang

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 rapid expansion of context window sizes in Large Language Models~(LLMs) has enabled them to tackle increasingly complex tasks involving lengthy documents. However, this progress comes at the cost of a substantial increase in memory…

Computation and Language · Computer Science 2025-08-05 Da Ma , Lu Chen , Situo Zhang , Yuxun Miao , Su Zhu , Zhi Chen , Hongshen Xu , Hanqi Li , Shuai Fan , Lei Pan , Kai Yu

Sparse Mixture-of-Experts (MoE) models scale capacity by routing each token to a small subset of experts. However, their routers exhibit a fundamental trade-off: strong load balancing can suppress expert specialization, while aggressive…

Machine Learning · Computer Science 2026-05-12 Gleb Molodtsov , Alexander Miasnikov , Aleksandr Beznosikov

Block-wise diffusion language models (DLMs) generate multiple tokens in any order, offering a promising alternative to the autoregressive decoding pipeline. However, they still remain bottlenecked by memory-bound attention in long-context…

Mixture-of-Experts (MoE) has emerged as a practical approach to scale up parameters for the Transformer model to achieve better generalization while maintaining a sub-linear increase in computation overhead. Current MoE models are mainly…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-04-03 Shuqing Luo , Jie Peng , Pingzhi Li , Hanrui Wang , Tianlong Chen

Long-context inference in large language models (LLMs) is bottlenecked by the linear growth of the self-attention key-value (KV) cache. Top-k sparse attention alleviates this by loading only a small fraction of the KV cache, but accurately…

Computation and Language · Computer Science 2026-05-28 Keqi Deng , Shaoshi Ling , Ruchao Fan , Jinyu Li

Scaling inference for large language models (LLMs) is increasingly constrained by limited GPU memory, especially due to growing key-value (KV) caches required for long-context generation. While existing approaches offload KV caches to CPU…

Machine Learning · Computer Science 2025-07-08 Weishu Deng , Yujie Yang , Peiran Du , Lingfeng Xiang , Zhen Lin , Chen Zhong , Song Jiang , Hui Lu , Jia Rao
‹ Prev 1 4 5 6 7 8 10 Next ›