English
Related papers

Related papers: Efficient Long-context Language Model Training by …

200 papers

Training causal transformers at extreme sequence lengths is bottlenecked by the quadratic time and memory of scaled dot-product attention (SDPA). In this work, we propose Lighthouse Attention, a training-only symmetrical selection-based…

Computation and Language · Computer Science 2026-05-08 Bowen Peng , Subho Ghosh , Jeffrey Quesnelle

Autoregressive (AR) generation is the standard decoding paradigm for Large Language Models (LLMs), but its token-by-token nature limits parallelism at inference time. Diffusion Language Models (DLLMs) offer parallel decoding by recovering…

Computation and Language · Computer Science 2025-12-30 Aiwei Liu , Minghua He , Shaoxun Zeng , Sijun Zhang , Linhao Zhang , Chuhan Wu , Wei Jia , Yuan Liu , Xiao Zhou , Jie Zhou

Masked diffusion language models (MDLMs) enable parallel decoding by predicting all masked positions at each denoising step, yet existing training-free samplers usually decide which positions to commit at token-level granularity. We revisit…

Machine Learning · Computer Science 2026-05-29 Heqiang Qi , Wei Huang , Mingyuan Bai , Xiangming Meng

The attention mechanism is an important reason for the success of transformers. It relies on computing pairwise relations between tokens. To reduce the high computational cost of standard quadratic attention, linear attention has been…

Artificial Intelligence · Computer Science 2026-02-13 Hanno Ackermann , Hong Cai , Mohsen Ghafoorian , Amirhossein Habibian

Large Language Models (LLMs) are increasingly deployed in long-context tasks such as reasoning, code generation, and multi-turn dialogue. However, inference over extended contexts is bottlenecked by the Key-Value (KV) cache, whose memory…

Computation and Language · Computer Science 2026-05-21 Seonghwan Choi , Beomseok Kang , Dongwon Jo , Jae-Joon Kim

Disaggregation has emerged as a powerful strategy for optimizing large language model (LLM) inference by separating compute-intensive prefill and memory-bound decode phases across specialized GPUs. This separation improves utilization and…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-21 Yiwei Jiang , Sangeeta Chowdhary , Nathaniel Morris , Rutwik Jain , Srilatha Manne , Sam Bayliss

The linear memory growth of the KV cache poses a significant bottleneck for LLM inference in long-context tasks. Existing static compression methods often fail to preserve globally important information. Although recent dynamic retrieval…

Computation and Language · Computer Science 2026-04-21 Zhiyuan Shi , Qibo Qiu , Feng Xue , Zhonglin Jiang , Li Yu , Jian Jiang , Xiaofei He , Wenxiao Wang

The use of complex attention modules has improved the performance of the Visual Question Answering (VQA) task. This work aims to learn an improved multi-modal representation through dense interaction of visual and textual modalities. The…

Computer Vision and Pattern Recognition · Computer Science 2023-03-01 Aakansha Mishra , Ashish Anand , Prithwijit Guha

In the framework of learned image compression, the context model plays a pivotal role in capturing the dependencies among latent representations. To reduce the decoding time resulting from the serial autoregressive context model, the…

Image and Video Processing · Electrical Eng. & Systems 2023-12-01 Yang Sui , Ding Ding , Xiang Pan , Xiaozhong Xu , Shan Liu , Bo Yuan , Zhenzhong Chen

Long-context inference increasingly operates over CPU-resident KV caches, either because decoding-time KV states exceed GPU memory capacity or because disaggregated prefill-decode systems place KV data in host memory. Although block-sparse…

Machine Learning · Computer Science 2026-05-11 Feiyu Yao , Zhixiong Niu , Xiaqing Li , Yongqiang Xiong , Juan Fang , Qian Wang

Multimodal in-context learning (ICL) is becoming a key capability that allows large vision-language models (LVLMs) to adapt to novel tasks without parameter updates, which expands their usefulness in many real-world applications. However,…

Computer Vision and Pattern Recognition · Computer Science 2025-12-11 Yanshu Li , Jianjiang Yang , Ziteng Yang , Bozheng Li , Ligong Han , Hongyang He , Zhengtao Yao , Yingjie Victor Chen , Songlin Fei , Dongfang Liu , Ruixiang Tang

The evolution of Large Language Models from the Transformer architecture to models with trillions of parameters has shifted the primary bottleneck from model training to real time inference. Deploying these massive models is a complex…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-12 Madabattula Rajesh Kumar , Srinivasa Rao Aravilli , Mustafa Saify , Shashank Srivastava

As Large Language Models (LLMs) scale to longer context windows, the computational cost of attention mechanisms, which traditionally grows quadratically with input length, presents a critical challenge for real-time and memory-constrained…

Computation and Language · Computer Science 2024-12-10 James Vo

Large language models have an exceptional capability to incorporate new information in a contextual manner. However, the full potential of such an approach is often restrained due to a limitation in the effective context length. One…

Computation and Language · Computer Science 2023-12-01 Szymon Tworkowski , Konrad Staniszewski , Mikołaj Pacek , Yuhuai Wu , Henryk Michalewski , Piotr Miłoś

Multi-head latent attention (MLA) is designed to optimize KV cache memory through low-rank key-value joint compression. Rather than caching keys and values separately, MLA stores their compressed latent representations, reducing memory…

Computation and Language · Computer Science 2025-09-09 Guihong Li , Mehdi Rezagholizadeh , Mingyu Yang , Vikram Appia , Emad Barsoum

Recently, conformer-based end-to-end automatic speech recognition, which outperforms recurrent neural network based ones, has received much attention. Although the parallel computing of conformer is more efficient than recurrent neural…

Sound · Computer Science 2021-07-26 Shengqiang Li , Menglong Xu , Xiao-Lei Zhang

When humans describe a visual scene, they do not process the entire image uniformly; instead, they selectively fixate on regions relevant to their intended description. In contrast, current multimodal large language models (MLLMs) attend to…

Computer Vision and Pattern Recognition · Computer Science 2026-05-14 Junha Song , Byeongho Heo , Geonmo Gu , Jaegul Choo , Dongyoon Han , Sangdoo Yun

Contextual information plays an important role in action recognition. Local operations have difficulty to model the relation between two elements with a long-distance interval. However, directly modeling the contextual information between…

Computer Vision and Pattern Recognition · Computer Science 2021-04-07 Congqi Cao , Yue Lu , Yifan Zhang , Dongmei Jiang , Yanning Zhang

Grouped-Query Attention (GQA) is a widely adopted strategy for reducing the computational cost of attention layers in large language models (LLMs). However, current GQA configurations are often suboptimal because they overlook how context…

Computation and Language · Computer Science 2025-09-29 Yingfa Chen , Yutong Wu , Chenyang Song , Zhen Leng Thai , Xingyu Shen , Xu Han , Zhiyuan Liu , Maosong Sun

Modern diffusion models, particularly those utilizing a Transformer-based UNet for denoising, rely heavily on self-attention operations to manage complex spatial relationships, thus achieving impressive generation performance. However, this…

Computer Vision and Pattern Recognition · Computer Science 2024-10-18 Songhua Liu , Weihao Yu , Zhenxiong Tan , Xinchao Wang