中文
相关论文

相关论文: On the Efficacy of Eviction Policy for Key-Value C…

200 篇论文

Recent advancements in Large Visual Language Models (LVLMs) have gained significant attention due to their remarkable reasoning capabilities and proficiency in generalization. However, processing a large number of visual tokens and…

计算机视觉与模式识别 · 计算机科学 2025-07-24 Kai Huang , Hao Zou , Bochen Wang , Ye Xi , Zhen Xie , Hao Wang

Early-Exit Large Language Models (EE-LLMs) enable high throughput inference by allowing tokens to exit early at intermediate layers. However, their throughput is limited by the computational and memory savings. Existing EE-LLM frameworks…

计算与语言 · 计算机科学 2025-11-03 Avinash Kumar , Shashank Nag , Jason Clemons , Lizy John , Poulami Das

Recent advances in large language models (LLMs) have showcased exceptional performance in long-context tasks, while facing significant inference efficiency challenges with limited GPU memory. Existing solutions first proposed the…

计算与语言 · 计算机科学 2025-02-20 Qingfa Xiao , Jiachuan Wang , Haoyang Li , Cheng Deng , Jiaqi Tang , Shuangyin Li , Yongqi Zhang , Jun Wang , Lei Chen

Large Language Model (LLM) serving is increasingly constrained by the growing size of the key-value (KV) cache, which scales with both context length and generation length. Prior work shows that attention is dominated by a small subset of…

机器学习 · 计算机科学 2026-04-21 Nazmul Takbir , Hamidreza Alikhani , Nikil Dutt , Sangeetha Abdu Jyothi

As LLMs have become capable of processing more complex types of inputs, researchers have recently studied how to efficiently and affordably process possibly arbitrarily long sequences. One effective approach is to use a FIFO memory to store…

计算与语言 · 计算机科学 2024-01-11 Zi Yang , Nan Hua

Large language models (LLMs) have demonstrated transformative capabilities across diverse artificial intelligence applications, yet their deployment is hindered by substantial memory and computational demands, especially in…

硬件体系结构 · 计算机科学 2025-05-13 Feng Cheng , Cong Guo , Chiyue Wei , Junyao Zhang , Changchun Zhou , Edward Hanson , Jiaqi Zhang , Xiaoxiao Liu , Hai "Helen" Li , Yiran Chen

Domain-specific large language models (LLMs), typically developed by fine-tuning a pre-trained general-purpose LLM on specialized datasets, represent a significant advancement in applied AI. A common strategy in LLM fine-tuning is…

Logical reasoning is a critical benchmark for evaluating the capabilities of large language models (LLMs), as it reflects their ability to derive valid conclusions from given premises. While the combination of test-time scaling with…

计算与语言 · 计算机科学 2025-08-28 Ramya Keerthy Thatikonda , Wray Buntine , Ehsan Shareghi

Autoregressive large language models (LLMs) have made remarkable progress in various natural language generation tasks. However, they incur high computation cost and latency resulting from the autoregressive token-by-token generation. To…

计算与语言 · 计算机科学 2023-07-07 Luciano Del Corro , Allie Del Giorno , Sahaj Agarwal , Bin Yu , Ahmed Awadallah , Subhabrata Mukherjee

Memory consumption of the Key-Value (KV) cache represents a major bottleneck for efficient large language model inference. While attention-score-based KV cache pruning shows promise, it faces critical practical limitations: attention scores…

人工智能 · 计算机科学 2025-10-02 Alessio Devoto , Maximilian Jeblick , Simon Jégou

Diffusion large language models (dLLMs) offer a promising route to parallel and efficient text generation, but improving their reasoning ability requires effective post-training. Reinforcement learning with verifiable rewards (RLVR) is a…

计算与语言 · 计算机科学 2026-05-12 Zichao Yu , Shengze Xu , Bingqing Jiang , Wenyi Zhang , Difan Zou

Large Reasoning Models (LRMs) significantly improve the reasoning ability of Large Language Models (LLMs) by learning to reason, exhibiting promising performance in solving complex tasks. However, their deliberative reasoning process leads…

Large Language Models (LLMs) are transforming recommendation from ranking into a generative task, but industrial deployment remains limited by the high latency of processing long, personalized prompts. Standard prefix caching provides…

分布式、并行与集群计算 · 计算机科学 2026-05-11 Zhan Zhao , Yuxin Wang , Amelie Chi Zhou

Large Language Models (LLMs) have achieved state-of-the-art performance in text re-ranking. This process includes queries and candidate passages in the prompts, utilizing pointwise, listwise, and pairwise prompting strategies. A limitation…

计算与语言 · 计算机科学 2024-05-29 Muhammad Shihab Rashid , Jannat Ara Meem , Yue Dong , Vagelis Hristidis

Interacting with humans through multi-turn conversations is a fundamental feature of large language models (LLMs). However, existing LLM serving engines executing multi-turn conversations are inefficient due to the need to repeatedly…

计算与语言 · 计算机科学 2024-07-02 Bin Gao , Zhuomin He , Puru Sharma , Qingxuan Kang , Djordje Jevdjic , Junbo Deng , Xingkun Yang , Zhou Yu , Pengfei Zuo

Continual learning (CL) is crucial for language models to dynamically adapt to the evolving real-world demands. To mitigate the catastrophic forgetting problem in CL, data replay has been proven a simple and effective strategy, and the…

计算与语言 · 计算机科学 2024-11-12 Jinghan He , Haiyun Guo , Kuan Zhu , Zihan Zhao , Ming Tang , Jinqiao Wang

Large Language Models (LLMs) require significant GPU memory when processing long texts, with the key value (KV) cache consuming up to 70\% of total memory during inference. Although existing compression methods reduce memory by evaluating…

计算与语言 · 计算机科学 2025-10-15 Xiang Liu , Zhenheng Tang , Peijie Dong , Zeyu Li , Yue Liu , Bo Li , Xuming Hu , Xiaowen Chu

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…

计算与语言 · 计算机科学 2026-04-09 Zhirui Chen , Peiyang Liu , Ling Shao

Key-Value (KV) cache plays a crucial role in accelerating inference in large language models (LLMs) by storing intermediate attention states and avoiding redundant computation during autoregressive generation. However, its memory footprint…

机器学习 · 计算机科学 2026-04-14 Yuzhen Mao , Qitong Wang , Martin Ester , Ke Li

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an effective approach for improving the reasoning abilities of large language models (LLMs). The Group Relative Policy Optimization (GRPO) family has demonstrated strong…

计算与语言 · 计算机科学 2025-11-10 Chenxi Liu , Junjie Liang , Yuqi Jia , Bochuan Cao , Yang Bai , Heng Huang , Xun Chen