English
Related papers

Related papers: Stream2LLM: Overlap Context Streaming and Prefill …

200 papers

Efficient execution of deep learning workloads on dataflow architectures is crucial for overcoming memory bottlenecks and maximizing performance. While streaming intermediate results between computation kernels can significantly improve…

Hardware Architecture · Computer Science 2025-09-24 Hanchen Ye , Deming Chen

This paper presents VideoStreaming, an advanced vision-language large model (VLLM) for video understanding, that capably understands arbitrary-length video with a constant number of video tokens streamingly encoded and adaptively selected.…

Computer Vision and Pattern Recognition · Computer Science 2024-05-28 Rui Qian , Xiaoyi Dong , Pan Zhang , Yuhang Zang , Shuangrui Ding , Dahua Lin , Jiaqi Wang

Multimodal Large Language Models (MLLMs) face significant computational overhead when processing long videos due to the massive number of visual tokens required. To improve efficiency, existing methods primarily reduce redundancy by pruning…

Artificial Intelligence · Computer Science 2026-05-22 Bingjun Luo , Tony Wang , Chaoqi Chen , Xinpeng Ding

Transformers suffer from a high computational cost that grows with sequence length for self-attention, making inference in long streams prohibited by memory consumption. Constant-memory alternatives such as RNNs and SSMs compress history…

Machine Learning · Computer Science 2026-04-24 Zhixin Zhang , Shabo Zhang , Chengcan Wu , Zeming Wei , Meng Sun

Large Language Models (LLMs) with expanding context windows face significant performance hurdles. While caching key-value (KV) states is critical for avoiding redundant computation, the storage footprint of long-context caches quickly…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-27 Zhiqiang Xie , Ziyi Xu , Mark Zhao , Yuwei An , Vikram Sharma Mailthody , Scott Mahlke , Michael Garland , Christos Kozyrakis

A well-known dilemma in large vision-language models (e.g., GPT-4, LLaVA) is that while increasing the number of vision tokens generally enhances visual understanding, it also significantly raises memory and computational costs, especially…

Computer Vision and Pattern Recognition · Computer Science 2024-08-30 Shiwei Wu , Joya Chen , Kevin Qinghong Lin , Qimeng Wang , Yan Gao , Qianli Xu , Tong Xu , Yao Hu , Enhong Chen , Mike Zheng Shou

Despite rapid advancements in lifelong learning (LLL) research, a large body of research mainly focuses on improving the performance in the existing \textit{static} continual learning (CL) setups. These methods lack the ability to succeed…

Machine Learning · Computer Science 2023-01-30 Soumya Banerjee , Vinay Kumar Verma , Vinay P. Namboodiri

Long-context Large Language Model (LLM) inference faces increasing compute bottlenecks as attention calculations scale with context length, primarily due to the growing KV-cache transfer overhead that saturates High Bandwidth Memory (HBM).…

Hardware Architecture · Computer Science 2025-08-13 Ming-Yen Lee , Faaiq Waqar , Hanchen Yang , Muhammed Ahosan Ul Karim , Harsono Simka , Shimeng Yu

KV cache restoration has emerged as a dominant bottleneck in serving long-context LLM workloads, including multi-turn conversations, retrieval-augmented generation, and agentic pipelines. Existing approaches treat restoration as a…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-29 Sean Nian , Jiahao Fang , Qilong Feng , Zhiyu Wu , Fan Lai

Large language models (LLMs) have been a disruptive innovation in recent years, and they play a crucial role in our daily lives due to their ability to understand and generate human-like text. Their capabilities include natural language…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-17 Akrit Mudvari , Yuang Jiang , Leandros Tassiulas

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

Meeting stringent Time-To-First-Token (TTFT) requirements is crucial for LLM applications. To improve efficiency, modern LLM serving systems adopt disaggregated architectures with diverse parallelisms, introducing complex multi-stage…

Networking and Internet Architecture · Computer Science 2026-03-19 Yijun Sun , Xudong Liao , Songrun Xie , Hao Chen , Han Tian , Wenxue Li , Yiming Zhang , Kai Chen

We study offline scheduling for large language model (LLM) serving under a fixed KV-cache memory budget, where requests have heterogeneous prompt (prefill) and response (decode) lengths. Prompt tokens determine initial KV usage, and each…

Optimization and Control · Mathematics 2026-02-11 Meixuan Wang , Yinyu Ye , Zijie Zhou

Occlusions between consecutive frames have long posed a significant challenge in optical flow estimation. The inherent ambiguity introduced by occlusions directly violates the brightness constancy constraint and considerably hinders…

Computer Vision and Pattern Recognition · Computer Science 2023-11-30 Shangkun Sun , Jiaming Liu , Thomas H. Li , Huaxia Li , Guoqing Liu , Wei Gao

Multimodal Large Language Models (MLLMs) have achieved strong performance across many tasks, yet most systems remain limited to offline inference, requiring complete inputs before generating outputs. Recent streaming methods reduce latency…

Computer Vision and Pattern Recognition · Computer Science 2026-01-13 Junyan Lin , Junlong Tong , Hao Wu , Jialiang Zhang , Jinming Liu , Xin Jin , Xiaoyu Shen

The limited context window of contemporary large language models (LLMs) remains a primary bottleneck for their broader application across diverse domains. Although continual pre-training on long-context data offers a straightforward…

Computation and Language · Computer Science 2026-04-10 Wei Han , Pan Zhou , Soujanya Poria , Shuicheng Yan

Large language model (LLM) inference serving systems are essential to various LLM-based applications. As demand for LLM services continues to grow, scaling these systems to handle high request rates while meeting latency Service-Level…

Machine Learning · Computer Science 2025-04-11 Shihong Gao , Xin Zhang , Yanyan Shen , Lei Chen

While Large Language Models (LLMs) have achieved remarkable success in various fields, the efficiency of training and inference remains a major challenge. To address this issue, we propose SUBLLM, short for Subsampling-Upsampling-Bypass…

Computation and Language · Computer Science 2024-08-26 Quandong Wang , Yuxuan Yuan , Xiaoyu Yang , Ruike Zhang , Kang Zhao , Wei Liu , Jian Luan , Daniel Povey , Bin Wang

Deploying million-token Large Language Models (LLMs) is challenging because production workloads are highly heterogeneous, mixing short queries and long documents. This heterogeneity, combined with the quadratic complexity of attention,…

The latency bottleneck of traditional text-to-speech (TTS) systems fundamentally hinders the potential of streaming large language models (LLMs) in conversational AI. These TTS systems, typically trained and inferenced on complete…

Computation and Language · Computer Science 2025-05-27 Richard He Bai , Zijin Gu , Tatiana Likhomanenko , Navdeep Jaitly