English
Related papers

Related papers: HieraSparse: Hierarchical Semi-Structured Sparse K…

200 papers

Scaling language models to long contexts is often bottlenecked by the size of the key-value (KV) cache. In deployed settings, long contexts are typically managed through compaction in token space via summarization. However, summarization…

Machine Learning · Computer Science 2026-05-28 Adam Zweiger , Xinghong Fu , Han Guo , Yoon Kim

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

The widespread of Large Language Models (LLMs) marks a significant milestone in generative AI. Nevertheless, the increasing context length and batch size in offline LLM inference escalate the memory requirement of the key-value (KV) cache,…

Hardware Architecture · Computer Science 2024-09-10 Xiurui Pan , Endian Li , Qiao Li , Shengwen Liang , Yizhou Shan , Ke Zhou , Yingwei Luo , Xiaolin Wang , Jie Zhang

Large Language Models (LLMs) have shown remarkable comprehension abilities but face challenges in GPU memory usage during inference, hindering their scalability for real-time applications like chatbots. To accelerate inference, we store…

Computation and Language · Computer Science 2024-06-06 Dongjie Yang , XiaoDong Han , Yan Gao , Yao Hu , Shilin Zhang , Hai Zhao

The key-value (KV) cache is a foundational optimization in Transformer-based large language models (LLMs), eliminating redundant recomputation of past token representations during autoregressive generation. However, its memory footprint…

Machine Learning · Computer Science 2026-03-24 Yichun Xu , Navjot K. Khaira , Tejinder Singh

KV cache compression is critical for efficient long-context LLM inference. Approaches that reduce the per-pair footprint -- quantization and low-rank decomposition -- are orthogonal to those that reduce the sequence length of the cache.…

Machine Learning · Computer Science 2026-03-31 Bo Jiang , Sian Jin

Generating long sequences of tokens given a long-context input is a very compute-intensive inference scenario for large language models (LLMs). One prominent inference speed-up approach is to construct a smaller key-value (KV) cache,…

Computation and Language · Computer Science 2025-03-04 Fangyuan Xu , Tanya Goyal , Eunsol Choi

Large language models (LLMs) now support context windows of hundreds of thousands to millions of tokens, enabling applications such as long-document summarization, large-scale code synthesis, multi-document question answering and persistent…

Computation and Language · Computer Science 2025-10-22 Siyuan Yan , Guo-Qing Jiang , Yuchen Zhang , Xiaoxing Ma , Ran Zhu , Chun Cao , Jingwei Xu

In long-context large language model (LLM) inference, the prefill stage dominates computation due to self-attention over the complete input context. Sparse attention significantly reduces self-attention computation by limiting each token's…

Hardware Architecture · Computer Science 2026-02-25 Rakshith Jayanth , Viktor Prasanna

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

The growing complexity of LLM usage today, e.g., multi-round conversation and retrieval-augmented generation (RAG), makes contextual states (i.e., KV cache) reusable across user requests. Given the capacity constraints of GPU memory, only a…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-08 Shiwei Gao , Youmin Chen , Jiwu Shu

The Key-Value (KV) cache introduces substantial memory overhead during large language model (LLM) inference. Although existing vector quantization (VQ) methods reduce KV cache usage and provide flexible representational capacity across…

Computation and Language · Computer Science 2025-10-08 Dingyu Yao , Chenxu Yang , Zhengyang Tong , Zheng Lin , Wei Liu , Jian Luan , Weiping Wang

Modern large language models (LLMs) extend context lengths to millions of tokens, enabling coherent, personalized responses grounded in long conversational history. However, the Key-Value (KV) cache grows linearly with the extended dialogue…

Computation and Language · Computer Science 2026-05-21 Minsoo Kim , Arnav Kundu , Han-Byul Kim , Richa Dixit , Minsik Cho

Generative large language models (LLMs) have revolutionized multiple domains. Modern LLMs predominantly rely on an autoregressive decoding strategy, which generates output tokens sequentially and employs a key-value cache (KV cache) to…

Cryptography and Security · Computer Science 2026-02-13 Ye Yu , Yifan Zhou , Yi Chen , Pedro Soto , Wenjie Xiong , Meng Li

Although Key-Value (KV) Cache is essential for efficient large language models (LLMs) inference, its growing memory footprint in long-context scenarios poses a significant bottleneck, making KVCache compression crucial. Current compression…

Machine Learning · Computer Science 2026-02-04 Jiancai Ye , Jun Liu , Qingchen Li , Tianlang Zhao , Hanbin Zhang , Jiayi Pan , Ningyi Xu , Guohao Dai

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

The growth of long-context Large Language Models (LLMs) significantly increases memory and bandwidth pressure during autoregressive decoding due to the expanding Key-Value (KV) cache. While accuracy-preserving KV-cache quantization (e.g.,…

Hardware Architecture · Computer Science 2026-01-06 Dayou Du , Shijie Cao , Jianyi Cheng , Luo Mai , Ting Cao , Mao Yang

The growing computational and memory demands of the Key-Value (KV) cache significantly limit the ability of Large Language Models (LLMs). While KV merging has emerged as a promising solution, existing methods that rely on empirical…

Computation and Language · Computer Science 2026-03-10 Lianjun Liu , Hongli An , Weiqi Yan , Xin Du , Shengchuan Zhang , Huazhong Liu , Yunshan Zhong

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

Efficient inference with Large Language Models (LLMs) increasingly relies on Key-Value (KV) caches to store previously computed key and value vectors at each layer. These caches are essential to minimize redundant computation during…

Hardware Architecture · Computer Science 2026-04-08 Oteo Mamo , Olga Kogiou , Hyunjin Yi , Weikuan Yu