English
Related papers

Related papers: ArborKV: Structure-Aware KV Cache Management for S…

200 papers

While Chain-of-Thought (CoT) prompting improves reasoning in large language models (LLMs), the excessive length of reasoning tokens increases latency and KV cache memory usage, and may even truncate final answers under context limits. We…

Computation and Language · Computer Science 2025-05-26 Gengyang Li , Yifeng Gao , Yuming Li , Yunfang Wu

Persistent key-value (KV) stores mostly build on the Log-Structured Merge (LSM) tree for high write performance, yet the LSM-tree suffers from the inherently high I/O amplification. KV separation mitigates I/O amplification by storing only…

Databases · Computer Science 2019-06-18 Yongkun Li , Helen H. W. Chan , Patrick P. C. Lee , Yinlong Xu

While inference-time scaling enables LLMs to carry out increasingly long and capable reasoning traces, the patterns and insights uncovered during these traces are immediately discarded once the context window is reset for a new query.…

Artificial Intelligence · Computer Science 2025-10-07 Matthew Ho , Chen Si , Zhaoxiang Feng , Fangxu Yu , Yichi Yang , Zhijian Liu , Zhiting Hu , Lianhui Qin

In this study, we investigate whether attention-based information flow inside large language models (LLMs) is aggregated through noticeable patterns for long context processing. Our observations reveal that LLMs aggregate information…

Computation and Language · Computer Science 2025-05-16 Zefan Cai , Yichi Zhang , Bofei Gao , Yuliang Liu , Yucheng Li , Tianyu Liu , Keming Lu , Wayne Xiong , Yue Dong , Junjie Hu , Wen Xiao

Multi-intent natural language understanding requires retrieval systems that simultaneously achieve high accuracy and computational efficiency, yet existing approaches apply either uniform single-step retrieval that compromises recall or…

Artificial Intelligence · Computer Science 2026-04-28 Hee-Kyong Yoo , Wonbae Kim , Hyocheol Ahn

Test-time compute scaling has emerged as a new axis along which to improve model accuracy, where additional computation is used at inference time to allow the model to think longer for more challenging problems. One promising approach for…

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

Large Language Models (LLMs) have demonstrated remarkable abilities across various language tasks, but solving complex reasoning problems remains a significant challenge. While existing methods, such as Chain-of-Thought (CoT) and…

Computation and Language · Computer Science 2025-04-02 Zhenni Bi , Kai Han , Chuanjian Liu , Yehui Tang , Yunhe Wang

Chain-of-thought reasoning has driven striking advances in language model capability, yet every reasoning step grows the KV cache, creating a bottleneck to scaling this paradigm further. Current approaches manage these constraints on the…

Machine Learning · Computer Science 2026-04-21 Michael Y. Li , Jubayer Ibn Hamid , Emily B. Fox , Noah D. Goodman

Large language models (LLMs) have been routinely used to solve various tasks using step-by-step reasoning. However, the structure of intermediate reasoning steps, or thoughts, is rigid and unidirectional, such as chains, trees, or…

Artificial Intelligence · Computer Science 2024-12-30 Sijia Chen , Baochun Li

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

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

While chain-of-thought (CoT) reasoning enables LLMs to solve challenging reasoning tasks, the linear growth of the KV cache leads to substantial memory and inference overhead. Existing approaches such as context compression and multi-token…

Artificial Intelligence · Computer Science 2026-05-29 Xinyu Liu , Xin Liu , Bo Jin , Runsong Zhao , Pengcheng Huang , Junhao Ruan , Bei Li , Chunyang Xiao , Chenglong Wang , Tong Xiao , Jingbo Zhu

Chain-of-thought (CoT) reasoning improves large language models (LLMs) on difficult tasks, but it also makes inference expensive because every intermediate step must be generated as a discrete token. Latent reasoning reduces visible token…

Computation and Language · Computer Science 2026-05-11 Xuan Li , Yining Wang , Yuchen Liu , Guanjun Liu , Delai Qiu , Shengping Liu , Jiaen Liang , Wei Huang , Jun Yu , Junnan Zhu

Memory-augmented Large Language Models (LLMs) have demonstrated remarkable capability for complex and long-horizon embodied planning. By keeping track of past experiences and environmental states, memory enables LLMs to maintain a global…

Robotics · Computer Science 2026-03-18 Zebin Yang , Tong Xie , Baotong Lu , Shaoshan Liu , Bo Yu , Meng Li

Tree-search decoding is an effective form of test-time scaling for large language models (LLMs), but real-world deployment imposes a fixed per-query token budget that varies across settings. Existing tree-search policies are largely…

Computation and Language · Computer Science 2026-02-11 Sora Miyamoto , Daisuke Oba , Naoaki Okazaki

The increasing size of the Key-Value (KV) cache during the Large Language Models long-context inference is the main obstacle for its balance between the deployment cost and task accuracy. To reduce the KV cache size in such scenarios, most…

Machine Learning · Computer Science 2025-07-25 Manlai Liang , JiaMing Zhang , Xiong Li , Jinlong Li

Multimodal Large Language Models (MLLMs) possess intrinsic reasoning and world-knowledge capabilities, yet adapting them for dense retrieval remains challenging. Existing approaches rely on invasive parameter updates, such as full…

Computer Vision and Pattern Recognition · Computer Science 2026-05-12 Haoran Lou , Ziyan Liu , Chunxiao Fan , Yuexin Wu , Yue Ming , Hao Wu , Kai Zuo , Yibo Chen , Xu Tang

We present Adaptive Soft Rolling KV Freeze with Entropy-Guided Recovery (ASR-KF-EGR), a training-free inference-time framework for efficient large language model generation. Our method introduces a reversible soft-freeze mechanism that…

Machine Learning · Computer Science 2025-12-15 Adilet Metinov , Gulida M. Kudakeeva , Bolotbek uulu Nursultan , Gulnara D. Kabaeva

Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications. However, their increased computational and memory demands present significant…

Computation and Language · Computer Science 2025-02-28 Yuhui Xu , Zhanming Jie , Hanze Dong , Lei Wang , Xudong Lu , Aojun Zhou , Amrita Saha , Caiming Xiong , Doyen Sahoo