English
Related papers

Related papers: TurboAngle: Near-Lossless KV Cache Compression via…

200 papers

Learned image compression has exhibited promising compression performance, but variable bitrates over a wide range remain a challenge. State-of-the-art variable rate methods compromise the loss of model performance and require numerous…

Image and Video Processing · Electrical Eng. & Systems 2023-03-13 Kedeng Tong , Yaojun Wu , Yue Li , Kai Zhang , Li Zhang , Xin Jin

Vector Quantization (VQ) is an appealing model compression method to obtain a tiny model with less accuracy loss. While methods to obtain better codebooks and codes under fixed clustering dimensionality have been extensively studied,…

Computer Vision and Pattern Recognition · Computer Science 2022-11-22 Zezhou Zhu , Yucong Zhou , Zhao Zhong

Inference-time scaling trades efficiency for increased reasoning accuracy by generating longer or more parallel sequences. However, in Transformer LLMs, generation cost is bottlenecked by the size of the key-value (KV) cache, rather than…

Machine Learning · Computer Science 2025-11-10 Adrian Łańcucki , Konrad Staniszewski , Piotr Nawrot , Edoardo M. Ponti

A key limitation of autoregressive Transformers is the large memory needed at inference-time to cache all previous key-value (KV) embeddings. Prior works address this by compressing the KV cache, but often assume specific structural…

Data Structures and Algorithms · Computer Science 2025-02-25 Themistoklis Haris , Krzysztof Onak

Autoregressive decoding in large language models (LLMs) requires caching a growing list of past key-value (KV) pairs, making long-context inference a memory-bound problem. While recent methods have explored quantizing the cache, evicting…

Computation and Language · Computer Science 2025-10-08 Harshil Vejendla

Post-training KV-Cache compression methods typically either sample a subset of effectual tokens or quantize the data into lower numerical bit width. However, these methods cannot exploit redundancy in the hidden dimension of the KV tensors.…

Latent variable models have been successfully applied in lossless compression with the bits-back coding algorithm. However, bits-back suffers from an increase in the bitrate equal to the KL divergence between the approximate posterior and…

This paper addresses the problem of lossy image compression, a fundamental problem in image processing and information theory that is involved in many real-world applications. We start by reviewing the framework of variational autoencoders…

Image and Video Processing · Electrical Eng. & Systems 2023-12-06 Zhihao Duan , Ming Lu , Jack Ma , Yuning Huang , Zhan Ma , Fengqing Zhu

Diffusion-based large language models (dLLMs) rely on bidirectional attention, which prevents lossless KV caching and requires a full forward pass at every denoising step. Existing approximate KV caching methods reduce this cost by…

Computation and Language · Computer Science 2026-03-20 Minsoo Cheong , Donghyun Son , Woosang Lim , Sungjoo Yoo

The development of large language models (LLMs) has significantly expanded model sizes, resulting in substantial GPU memory requirements during inference. The key and value storage of the attention map in the KV (key-value) cache accounts…

Machine Learning · Computer Science 2024-10-25 Yifei Yang , Zouying Cao , Qiguang Chen , Libo Qin , Dongjie Yang , Hai Zhao , Zhi Chen

Withtherapid advancement of large language models (LLMs), the context length for inference has been continuously increasing, leading to an exponential growth in the demand for Key-Value (KV) caching. This has resulted in a significant…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-11 Yanyu Liu , Jingying Fu , Sixiang Liu , Yitian Zou , You Fu , Jiehan Zhou , Shouhua Zhang

KV caching is a ubiquitous optimization in autoregressive transformer inference, long presumed to be numerically equivalent to cache-free computation. This assumption fails under standard FP16 precision: cache-ON and cache-OFF execution…

Machine Learning · Computer Science 2026-04-20 Ranjith Chodavarapu , Lei Xu

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

Transformer-based Large Language Models rely critically on the KV cache to efficiently handle extended contexts during the decode phase. Yet, the size of the KV cache grows proportionally with the input length, burdening both memory…

Computation and Language · Computer Science 2025-08-14 Payman Behnam , Yaosheng Fu , Ritchie Zhao , Po-An Tsai , Zhiding Yu , Alexey Tumanov

Autoregressive "language" models (LMs) trained on raw waveforms can be repurposed for lossless audio compression, but prior work is limited to 8-bit audio, leaving open whether such approaches work for practical settings (16/24-bit) and can…

Sound · Computer Science 2026-03-10 Phillip Long , Zachary Novack , Chris Donahue

The benefits of most large language models come with steep and often hidden economic and environmental costs due to their resource usage inefficiency during deployment. Model quantization improves energy and memory efficiency through…

Machine Learning · Computer Science 2026-01-14 Deyu Cao , Yixin Yin , Samin Aref

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

Deep image compression systems mainly contain four components: encoder, quantizer, entropy model, and decoder. To optimize these four components, a joint rate-distortion framework was proposed, and many deep neural network-based methods…

Image and Video Processing · Electrical Eng. & Systems 2020-07-27 Zhisheng Zhong , Hiroaki Akutsu , Kiyoharu Aizawa

The quadratic computational complexity of the standard attention mechanism constitutes a fundamental bottleneck for large language models in long-context inference. While existing KV cache compression methods alleviate memory pressure, they…

Computation and Language · Computer Science 2026-05-06 Jinyu Guo , Zhihan Zhang , Jiehui Xie , Md. Tamim Iqbal , Dongshen Han , Lik-Hang Lee , Sung-Ho Bae , Jie Zou , Yang Yang , Chaoning Zhang

Recent advances in deep generative modeling have enabled efficient modeling of high dimensional data distributions and opened up a new horizon for solving data compression problems. Specifically, autoencoder based learned image or video…

Machine Learning · Computer Science 2020-04-10 Adam Golinski , Reza Pourreza , Yang Yang , Guillaume Sautiere , Taco S Cohen