English
Related papers

Related papers: Compress, Gather, and Recompute: REFORMing Long-Co…

200 papers

Transformers have a quadratic scaling of computational complexity with input size, which limits the input context window size of large language models (LLMs) in both training and inference. Meanwhile, retrieval-augmented generation (RAG)…

Computation and Language · Computer Science 2024-10-18 Yimin Tang , Yurong Xu , Ning Yan , Masood Mortazavi

While the Mamba architecture demonstrates superior inference efficiency and competitive performance on short-context natural language processing (NLP) tasks, empirical evidence suggests its capacity to comprehend long contexts is limited…

Computation and Language · Computer Science 2025-01-03 Danlong Yuan , Jiahao Liu , Bei Li , Huishuai Zhang , Jingang Wang , Xunliang Cai , Dongyan Zhao

A recent trend in LLMs is developing recurrent sub-quadratic models that improve long-context processing efficiency. We investigate leading large long-context models, focusing on how their fixed-size recurrent memory affects their…

Machine Learning · Computer Science 2025-09-10 Assaf Ben-Kish , Itamar Zimerman , M. Jehanzeb Mirza , Lior Wolf , James Glass , Leonid Karlinsky , Raja Giryes

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

Retrieval-Augmented Generation (RAG) allows overcoming the limited knowledge of LLMs by extending the input with external information. As a consequence, the contextual inputs to the model become much longer which slows down decoding time…

Computation and Language · Computer Science 2024-10-30 David Rau , Shuai Wang , Hervé Déjean , Stéphane Clinchant

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

Scaling language models to longer contexts is essential for capturing rich dependencies across extended discourse. However, na\"ive context extension imposes significant computational and memory burdens, often resulting in inefficiencies…

Computation and Language · Computer Science 2026-02-03 Wenhao Li , Bangcheng Sun , Weihao Ye , Tianyi Zhang , Daohai Yu , Fei Chao , Rongrong Ji

Transformer-based large language models (LLMs) demonstrate impressive performance across various natural language processing tasks. Serving LLM inference for generating long contents, however, poses a challenge due to the enormous memory…

Machine Learning · Computer Science 2024-07-01 Wonbeom Lee , Jungi Lee , Junghwan Seo , Jaewoong Sim

Large Transformer models routinely achieve state-of-the-art results on a number of tasks but training these models can be prohibitively costly, especially on long sequences. We introduce two techniques to improve the efficiency of…

Machine Learning · Computer Science 2020-02-19 Nikita Kitaev , Łukasz Kaiser , Anselm Levskaya

Large language models (LLMs) based on Transformer have been widely applied in the filed of natural language processing (NLP), demonstrating strong performance, particularly in handling short text tasks. However, when it comes to long…

Computation and Language · Computer Science 2025-07-09 Yijun Liu , Jinzheng Yu , Yang Xu , Zhongyang Li , Qingfu Zhu

The transformer's context window is vital for tasks such as few-shot learning and conditional generation as it preserves previous tokens for active memory. However, as the context lengths increase, the computational costs grow…

Computation and Language · Computer Science 2025-04-01 Jeffrey Willette , Heejun Lee , Youngwan Lee , Myeongjae Jeon , Sung Ju Hwang

Autoregressive Transformers adopted in Large Language Models (LLMs) are hard to scale to long sequences. Despite several works trying to reduce their computational cost, most of LLMs still adopt attention layers between all pairs of tokens…

Computation and Language · Computer Science 2024-06-03 Sotiris Anagnostidis , Dario Pavllo , Luca Biggio , Lorenzo Noci , Aurelien Lucchi , Thomas Hofmann

The quadratic complexity of standard self-attention severely limits the application of Transformer-based models to long-context tasks. While efficient Transformer variants exist, they often require architectural changes and costly…

Computation and Language · Computer Science 2025-11-14 Jiangshu Du , Wenpeng Yin , Philip Yu

AI applications increasingly depend on long-context inference, where LLMs consume substantial context to support stronger reasoning. Common examples include retrieval-augmented generation, agent memory layers, and multi-agent orchestration.…

Machine Learning · Computer Science 2026-05-07 Yinsicheng Jiang , Yeqi Huang , Liang Cheng , Cheng Deng , Xuan Sun , Luo Mai

Sequence-to-sequence tasks often benefit from long contexts, but the quadratic complexity of self-attention in standard Transformers renders this non-trivial. During generation, temporary representations -stored in the so-called KV…

Computation and Language · Computer Science 2025-03-14 Vivek Chari , Guanghui Qin , Benjamin Van Durme

Repository-level code intelligence tasks require large language models (LLMs) to process long, multi-file contexts. Such inputs introduce three challenges: crucial context can be obscured by noise, truncated due to limited windows, and…

Software Engineering · Computer Science 2026-04-16 Jia Feng , Zhanyue Qin , Cuiyun Gao , Ruiqi Wang , Chaozheng Wang , Yingwei Ma , Xiaoyuan Xie

Large language models face challenges in long-context question answering, where key evidence of a query may be dispersed across millions of tokens. Existing works equip large language models with a memory buffer that is dynamically updated…

Computation and Language · Computer Science 2026-03-03 Yaorui Shi , Yuxin Chen , Siyuan Wang , Sihang Li , Hengxing Cai , Qi Gu , Xiang Wang , An Zhang

Large language models (LLMs) demonstrate exceptional capabilities in various scenarios. However, they suffer from much redundant information and are sensitive to the position of key information in long context scenarios. To address these…

Computation and Language · Computer Science 2025-02-11 Jiwei Tang , Jin Xu , Tingwei Lu , Zhicheng Zhang , Yiming Zhao , Lin Hai , Hai-Tao Zheng

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

Long context compression is a critical research problem due to its significance in reducing the high computational and memory costs associated with LLMs. In this paper, we propose Activation Beacon, a plug-in module for transformer-based…

Computation and Language · Computer Science 2024-10-14 Peitian Zhang , Zheng Liu , Shitao Xiao , Ninglu Shao , Qiwei Ye , Zhicheng Dou