English
Related papers

Related papers: Tensor Cache: Eviction-conditioned Associative Mem…

200 papers

KV cache quantization reduces the memory cost of long-context LLM inference, but introduces approximation error that is typically validated only empirically. Existing systems rely on average-case robustness, with no mechanism to detect or…

Machine Learning · Computer Science 2026-05-21 Dean Calver

The growing size of Large Language Models (LLMs) makes efficient inference challenging, primarily due to the memory demands of the autoregressive Key-Value (KV) cache. Existing eviction or compression methods reduce cost but rely on…

Computation and Language · Computer Science 2026-02-12 Luca Moschella , Laura Manduchi , Ozan Sener

The quadratic complexity of the attention module makes it gradually become the bulk of compute in Transformer-based LLMs during generation. Moreover, the excessive key-value cache that arises when dealing with long inputs also brings severe…

Computation and Language · Computer Science 2023-10-17 Siyu Ren , Qi Jia , Kenny Q. Zhu

Multimodal Large Language Models face severe challenges in computational efficiency and memory consumption due to the substantial expansion of the visual KV cache when processing long visual contexts. Existing KV cache compression methods…

Machine Learning · Computer Science 2026-05-07 Sihao Liu , YuFan Xiong , Zhonghua Jiang , Zhaode Wang , chengfei lv Shengyu Zhang

Efficient key-value (KV) cache management is crucial for the practical deployment of large language models (LLMs), yet existing compression techniques often incur a trade-off between performance degradation and computational overhead. We…

Machine Learning · Computer Science 2026-02-10 Jang-Hyun Kim , Dongyoon Han , Sangdoo Yun

Several works have developed eviction policies to remove key-value (KV) pairs from the KV cache for more efficient inference. The focus has been on compressing the KV cache after the input prompt has been processed for faster token…

Computation and Language · Computer Science 2025-07-04 Michael R. Metel , Boxing Chen , Mehdi Rezagholizadeh

Transformer-based large language models (LLMs) encounter challenges in processing long sequences on edge devices due to the quadratic complexity of attention mechanisms and growing memory demands from Key-Value (KV) cache. Existing KV cache…

Computation and Language · Computer Science 2025-03-31 Jiyu Chen , Shuang Peng , Daxiong Luo , Fan Yang , Renshou Wu , Fangyuan Li , Xiaoxin Chen

Large language models are increasingly applied to multi-document and long-form inputs, yet long-context inference remains memory- and noise-inefficient. Key-value (KV) caching scales linearly with context length, while external retrieval…

Computation and Language · Computer Science 2026-01-29 Qingsen Ma , Dianyun Wang , Yaoye Wang , Lechen Ning , Sujie Zhu , Xiaohang Zhang , Jiaming Lyu , Linhao Ren , Zhenbo Xu , Zhaofeng He

Slim attention shrinks the context memory size by 2x for transformer models with MHA (multi-head attention), which can speed up inference by up to 2x for large context windows. Slim attention is an exact, mathematically identical…

Machine Learning · Computer Science 2025-06-04 Nils Graef , Andrew Wasielewski

Incorporating external knowledge in large language models (LLMs) enhances their utility across diverse applications, but existing methods have trade-offs. Retrieval-Augmented Generation (RAG) fetches evidence via similarity search, but key…

Computation and Language · Computer Science 2025-03-10 Giulio Corallo , Orion Weller , Fabio Petroni , Paolo Papotti

Large language models (LLMs) based on Transformer Decoders have become the preferred choice for conversational generative AI. Despite the overall superiority of the Decoder architecture, the gradually increasing Key-Value (KV) cache during…

Computation and Language · Computer Science 2025-07-16 Luohe Shi , Zuchao Li , Lefei Zhang , Guoming Liu , Baoyuan Qi , Hai Zhao

As Large Language Models (LLMs) scale to support context windows exceeding one million tokens, the linear growth of Key-Value (KV) cache imposes severe memory capacity and bandwidth bottlenecks, constraining the efficiency of long-context…

Computation and Language · Computer Science 2026-04-09 Zhirui Chen , Peiyang Liu , Ling Shao

Autoregressive video diffusion models enable streaming generation, opening the door to long-form synthesis, video world models, and interactive neural game engines. However, their core attention layers become a major bottleneck at inference…

Computer Vision and Pattern Recognition · Computer Science 2026-02-03 Dvir Samuel , Issar Tzachor , Matan Levy , Micahel Green , Gal Chechik , Rami Ben-Ari

Large language models (LLMs) can now handle longer sequences of tokens, enabling complex tasks like book understanding and generating lengthy novels. However, the key-value (KV) cache required for LLMs consumes substantial memory as context…

Machine Learning · Computer Science 2024-11-13 Haojie Duanmu , Zhihang Yuan , Xiuhong Li , Jiangfei Duan , Xingcheng Zhang , Dahua Lin

Although the Transformer has become the cornerstone of modern AI, its autoregressive inference suffers from a linearly growing KV Cache and a computational complexity of O(N^2 d), severely hindering its ability to process ultra-long…

Machine Learning · Computer Science 2025-09-03 Zhongpan Tang

Transformers have emerged as the underpinning architecture for Large Language Models (LLMs). In generative language models, the inference process involves two primary phases: prompt processing and token generation. Token generation, which…

Machine Learning · Computer Science 2024-04-09 Muhammad Adnan , Akhil Arunkumar , Gaurav Jain , Prashant J. Nair , Ilya Soloveychik , Purushotham Kamath

Serving transformer language models with high throughput requires caching Key-Values (KVs) to avoid redundant computation during autoregressive generation. The memory footprint of KV caching is significant and heavily impacts serving costs.…

Machine Learning · Computer Science 2026-04-28 Anastasiia Filippova , David Grangier , Marco Cuturi , João Monteiro

Recent work has shown that attention-based language models excel at recall, the ability to ground generations in tokens previously seen in context. However, the efficiency of attention-based models is bottle-necked during inference by the…

Computation and Language · Computer Science 2025-03-10 Simran Arora , Sabri Eyuboglu , Michael Zhang , Aman Timalsina , Silas Alberti , Dylan Zinsley , James Zou , Atri Rudra , Christopher Ré

Large language models (LLMs) rely on key-value (KV) caches for efficient autoregressive decoding; however, cache size grows linearly with context length and model depth, becoming a major bottleneck in long-context inference. Prior KV cache…

Machine Learning · Computer Science 2025-09-22 Dmitry Akulov , Mohamed Sana , Antonio De Domenico , Tareq Si Salem , Nicola Piovesan , Fadhel Ayed

While Transformer self-attention offers strong parallelism, the Key-Value (KV) cache grows linearly with sequence length and becomes a bottleneck for inference efficiency. Multi-head latent attention was recently developed to compress the…

Machine Learning · Computer Science 2025-11-04 Keqi Deng , Philip C. Woodland