English
Related papers

Related papers: ProxyKV: Cross-Model Proxy Pruning for Efficient L…

200 papers

Despite their powerful capabilities, Multimodal Large Language Models (MLLMs) suffer from considerable computational overhead due to their reliance on massive visual tokens. Recent studies have explored token pruning to alleviate this…

Computer Vision and Pattern Recognition · Computer Science 2025-10-13 Xin Zou , Di Lu , Yizhou Wang , Yibo Yan , Yuanhuiyi Lyu , Xu Zheng , Linfeng Zhang , Xuming Hu

Modern language models are trained almost exclusively on token sequences produced by a fixed tokenizer, an external lossless compressor often over UTF-8 byte sequences, thereby coupling the model to that compressor. This work introduces…

Computation and Language · Computer Science 2026-05-15 Lin Zheng , Xinyu Li , Qian Liu , Xiachong Feng , Lingpeng Kong

Large language models (LLMs) achieve state-of-the-art accuracy on complex reasoning tasks by generating multiple chain-of-thought (CoT) traces, but using a fixed token budget per query leads to over-computation on easy inputs and…

Artificial Intelligence · Computer Science 2026-02-03 Katrina Brown , Aneesh Muppidi , Rana Shahout

With the advancements in long-context inference capabilities of large language models (LLMs), the KV cache has become one of the foundational components. However, its substantial GPU memory consumption makes KV cache compression a key…

Computation and Language · Computer Science 2025-03-28 Youhui Zuo , Sibo Wei , Chen Zhang , Zhuorui Liu , Wenpeng Lu , Dawei Song

Vision-Language Large Models (VLLMs) face significant efficiency challenges when processing high-resolution inputs. The quadratic complexity in attention and autoregressive generation, as well as the constantly growing key value (KV) cache…

Multimedia · Computer Science 2025-10-31 Zhonghua Jiang , Kunxi Li , Yiyun Zhou , Sihao Liu , Zhaode Wang , Chengfei lv , Shengyu Zhang

Generative Large Language Models (LLMs) have demonstrated remarkable results for a wide range of tasks. However, deploying these models for inference has been a significant challenge due to their unprecedented resource requirements. This…

Computation and Language · Computer Science 2024-06-06 Sehoon Kim , Coleman Hooper , Amir Gholami , Zhen Dong , Xiuyu Li , Sheng Shen , Michael W. Mahoney , Kurt Keutzer

How to efficiently serve LLMs in practice has become exceptionally challenging due to their prohibitive memory and computation requirements. In this study, we investigate optimizing the KV cache, whose memory footprint poses a critical…

Computation and Language · Computer Science 2025-06-10 Akshat Sharma , Hangliang Ding , Jianping Li , Neel Dani , Minjia Zhang

Long-horizon tasks requiring multi-step reasoning and dynamic re-planning remain challenging for large language models (LLMs). Sequential prompting methods are prone to context drift, loss of goal information, and recurrent failure cycles,…

Artificial Intelligence · Computer Science 2025-10-30 Zhenyu Zhang , Tianyi Chen , Weiran Xu , Alex Pentland , Jiaxin Pei

Multimodal Large Language Models (MLLMs) hold huge potential for usage in the medical domain, but their computational costs necessitate efficient compression techniques. This paper evaluates the impact of structural pruning and…

Artificial Intelligence · Computer Science 2025-09-25 Tanvir A. Khan , Aranya Saha , Ismam N. Swapnil , Mohammad A. Haque

Deploying long-context large language models (LLMs) is essential but poses significant computational and memory challenges. Caching all Key and Value (KV) states across all attention heads consumes substantial memory. Existing KV cache…

Computation and Language · Computer Science 2024-10-15 Guangxuan Xiao , Jiaming Tang , Jingwei Zuo , Junxian Guo , Shang Yang , Haotian Tang , Yao Fu , Song Han

Video large language models (video LLMs) excel at video comprehension but face significant computational inefficiency due to redundant video tokens. Existing token pruning methods offer solutions. However, approaches operating within the…

Computer Vision and Pattern Recognition · Computer Science 2025-10-13 Kele Shao , Keda Tao , Can Qin , Haoxuan You , Yang Sui , Huan Wang

Long-horizon language agents accumulate conversation history far faster than any fixed context window can hold, making memory management critical to both answer accuracy and serving cost. Existing approaches either expand the context window…

Computation and Language · Computer Science 2026-05-25 Jingyi Peng , Zhongwei Wan , Weiting Liu , Qiuzhuang Sun

Large Language Models (LLMs) suffer inference-time memory bottlenecks dominated by the attention Key-Value (KV) cache, which scales with model size and context length. While KV-cache quantization alleviates this cost, bit allocation between…

Machine Learning · Computer Science 2026-05-12 Mohsen Hariri , Alan Luo , Weicong Chen , Shaochen Zhong , Tianyi Zhang , Qifan Wang , Xia Hu , Xiaotian Han , Vipin Chaudhary

Vision-Language Models (VLMs) face severe memory and latency bottlenecks due to high-resolution visual tokens. While current token reduction methods theoretically save FLOPs, post-hoc pruning introduces structural overhead, failing to yield…

Artificial Intelligence · Computer Science 2026-05-28 Fengze Yang , Bo Yu , Xuewen Luo , Cathy Liu , Chenxi Liu

With the development of large language models (LLMs), efficient inference through Key-Value (KV) cache compression has attracted considerable attention, especially for long-context generation. To compress the KV cache, recent methods…

Computation and Language · Computer Science 2025-10-28 Qingyue Yang , Jie Wang , Xing Li , Zhihai Wang , Chen Chen , Lei Chen , Xianzhi Yu , Wulong Liu , Jianye Hao , Mingxuan Yuan , Bin Li

In this work, we present FastAV, the first token pruning framework tailored for audio-visual large language models (AV-LLMs). While token pruning has been actively explored in standard large language models (LLMs) and vision-language models…

Machine Learning · Computer Science 2026-01-21 Chaeyoung Jung , Youngjoon Jang , Seungwoo Lee , Joon Son Chung

Large language models (LLMs) are essential in natural language processing but often struggle with inference speed and computational efficiency, limiting real-time deployment. The key-value (KV) cache mechanism reduces computational overhead…

Computation and Language · Computer Science 2024-10-31 Junqi Zhao , Zhijin Fang , Shu Li , Shaohui Yang , Shichao He

Transformer-based Large Language Models (LLMs) have significantly advanced AI capabilities but pose considerable challenges for deployment on edge devices due to high computational demands, memory bandwidth constraints, and energy…

Hardware Architecture · Computer Science 2025-04-25 Maoyang Xiang , Ramesh Fernando , Bo Wang

Large language models (LLMs) demonstrate remarkable capabilities but face substantial serving costs due to their high memory demands, with the key-value (KV) cache being a primary bottleneck. State-of-the-art KV cache compression…

Machine Learning · Computer Science 2025-09-03 Yanqi Zhang , Yuwei Hu , Runyuan Zhao , John C. S. Lui , Haibo Chen

KV cache eviction has emerged as an effective solution to alleviate resource constraints faced by LLMs in long-context scenarios. However, existing token-level eviction methods often overlook two critical aspects: (1) their irreversible…

Machine Learning · Computer Science 2026-01-21 Yi Zhao , Yajuan Peng , Cam-Tu Nguyen , Zuchao Li , Xiaoliang Wang , Hai Zhao , Xiaoming Fu