English
Related papers

Related papers: Mixture of Hidden-Dimensions Transformer

200 papers

LLM architecture research generally aims to maximize model quality subject to fixed compute/latency budgets. However, many applications of interest such as edge and on-device deployment are further constrained by the model's memory…

Machine Learning · Computer Science 2026-04-28 Abbas Zeitoun , Lucas Torroba-Hennigen , Yoon Kim

The escalating size of Mixture-of-Experts (MoE) based Large Language Models (LLMs) presents significant computational and memory challenges, necessitating innovative solutions to enhance efficiency without compromising model accuracy.…

Machine Learning · Computer Science 2025-03-17 Chenpeng Wu , Qiqi Gu , Heng Shi , Jianguo Yao , Haibing Guan

Transformers allow attention between all pairs of tokens, but there is reason to believe that most of these connections - and their quadratic time and memory - may not be necessary. But which ones? We evaluate the impact of sparsification…

Computation and Language · Computer Science 2022-10-11 Siddhartha Brahma , Polina Zablotskaia , David Mimno

Natural human interactions for Mixed Reality Applications are overwhelmingly multimodal: humans communicate intent and instructions via a combination of visual, aural and gestural cues. However, supporting low-latency and accurate…

Human-Computer Interaction · Computer Science 2020-12-21 Darshana Rathnayake , Ashen de Silva , Dasun Puwakdandawa , Lakmal Meegahapola , Archan Misra , Indika Perera

Residual connections are central to modern deep learning architectures, enabling the training of very deep networks by mitigating gradient vanishing. Hyper-Connections recently generalized residual connections by introducing multiple…

Machine Learning · Computer Science 2025-03-19 Defa Zhu , Hongzhi Huang , Jundong Zhou , Zihao Huang , Yutao Zeng , Banggu Wu , Qiyang Min , Xun Zhou

With the success of language pretraining, it is highly desirable to develop more efficient architectures of good scalability that can exploit the abundant unlabeled data at a lower cost. To improve the efficiency, we examine the…

Machine Learning · Computer Science 2020-06-08 Zihang Dai , Guokun Lai , Yiming Yang , Quoc V. Le

A variable-phase-shifter (VPS) architecture with hybrid precoding for mixture use of phase shifters and switches, is proposed for millimeter wave massive multiple-input multiple-output communications. For the VPS architecture, a hybrid…

Information Theory · Computer Science 2022-08-02 Chenhao Qi , Qiang Liu , Xianghao Yu , Geoffrey Ye Li

Large Transformer models yield impressive results on many tasks, but are expensive to train, or even fine-tune, and so slow at decoding that their use and study becomes out of reach. We address this problem by leveraging sparsity. We study…

Transformer architectures dominate modern NLP but often demand heavy computational resources and intricate hyperparameter tuning. To mitigate these challenges, we propose a novel framework, BoostTransformer, that augments transformers with…

Machine Learning · Computer Science 2025-11-04 Biyi Fang , Truong Vo , Jean Utke , Diego Klabjan

Tensors play a vital role in machine learning (ML) and often exhibit properties best explored while maintaining high-order. Efficiently performing ML computations requires taking advantage of sparsity, but generalized hardware support is…

Hardware Architecture · Computer Science 2024-04-26 Gabriel Kulp , Andrew Ensinger , Lizhong Chen

Masked Discrete Diffusion Models (MDMs) have achieved strong performance across a wide range of multimodal tasks, including image understanding, generation, and editing. However, their inference speed remains suboptimal due to the need to…

Computer Vision and Pattern Recognition · Computer Science 2025-12-17 Shufan Li , Jiuxiang Gu , Kangning Liu , Zhe Lin , Zijun Wei , Aditya Grover , Jason Kuen

Transformer-based architectures have shown remarkable performance in vision and language tasks but pose unique challenges for safety-critical applications. This paper presents a conceptual framework for integrating Transformers into…

Software Engineering · Computer Science 2026-01-28 Sven Kirchner , Nils Purschke , Chengdong Wu , Alois Knoll

We present Token-UNet, adopting the TokenLearner and TokenFuser modules to encase Transformers into UNets. While Transformers have enabled global interactions among input elements in medical imaging, current computational challenges hinder…

Computer Vision and Pattern Recognition · Computer Science 2026-02-24 Louis Fabrice Tshimanga , Andrea Zanola , Federico Del Pup , Manfredo Atzori

An important development in deep learning from the earliest MLPs has been a move towards architectures with structural inductive biases which enable the model to keep distinct sources of information and routes of processing well-separated.…

Machine Learning · Computer Science 2021-03-02 Alex Lamb , Di He , Anirudh Goyal , Guolin Ke , Chien-Feng Liao , Mirco Ravanelli , Yoshua Bengio

We propose MUltiway Dynamic Dense (MUDD) connections, a simple yet effective method to address the limitations of residual connections and enhance cross-layer information flow in Transformers. Unlike existing dense connection approaches…

Machine Learning · Computer Science 2025-05-29 Da Xiao , Qingye Meng , Shengping Li , Xingyuan Yuan

With the rise and ubiquity of larger deep learning models, the need for high-quality compression techniques is growing in order to deploy these models widely. The sheer parameter count of these models makes it difficult to fit them into the…

Computation and Language · Computer Science 2025-04-01 Neha Verma , Kenton Murray , Kevin Duh

Despite their practical success, it remains unclear why Mixture of Experts (MoE) models can outperform dense networks beyond sheer parameter scaling. We study an iso-parameter regime where inputs exhibit latent modular structure but are…

Machine Learning · Computer Science 2026-01-22 Dong Sun , Rahul Nittala , Rebekka Burkholz

Mixture-of-Experts large language models (MoE-LLMs) marks a significant step forward of language models, however, they encounter two critical challenges in practice: 1) expert parameters lead to considerable memory consumption and loading…

Machine Learning · Computer Science 2025-02-25 Wei Huang , Yue Liao , Jianhui Liu , Ruifei He , Haoru Tan , Shiming Zhang , Hongsheng Li , Si Liu , Xiaojuan Qi

Previous work has demonstrated that MLPs within ReLU Transformers exhibit high levels of sparsity, with many of their activations equal to zero for any given token. We build on that work to more deeply explore how token-level sparsity…

Machine Learning · Computer Science 2024-07-11 Cody Wild , Jesper Anderson

Transformer architectures have achieved remarkable success across language, vision, and multimodal tasks, and there is growing demand for them to address in-context compositional learning tasks. In these tasks, models solve the target…

Machine Learning · Computer Science 2025-11-26 Wei Chen , Jingxi Yu , Zichen Miao , Qiang Qiu