English
Related papers

Related papers: Sparsity Moves Computation: How FFN Architecture R…

200 papers

Sparse Mixture of Experts (MoE) models offer a scalable and efficient architecture for training large neural networks by activating only a subset of parameters ("experts") for each input. A learned router computes a distribution over these…

Machine Learning · Computer Science 2025-10-14 Nabil Omi , Siddhartha Sen , Ali Farhadi

Graph Transformers (GTs) have demonstrated superior performance compared to traditional message-passing graph neural networks in many studies, especially in processing graph data with long-range dependencies. However, GTs tend to suffer…

Machine Learning · Computer Science 2025-04-30 Zhonghao Li , Ji Shi , Xinming Zhang , Miao Zhang , Bo Li

Packing is a crucial step of FPGA design, directly impacting interconnect complexity, routing congestion, and overall performance. This paper presents a post-packing interconnect-aware analysis, illustrating how dense (sparse) packing…

Systems and Control · Electrical Eng. & Systems 2025-04-11 X. Wang , D. Stroobandt

Transformers have achieved remarkable success across diverse domains, but their monolithic architecture presents challenges in interpretability, adaptability, and scalability. This paper introduces a novel modular Transformer architecture…

Machine Learning · Computer Science 2025-01-07 Zhenyu Guo , Wenguang Chen

To alleviate the computational burden of large language models (LLMs), architectures with activation sparsity, represented by mixture-of-experts (MoE), have attracted increasing attention. However, the non-differentiable and inflexible…

Machine Learning · Computer Science 2025-07-31 Chenyang Song , Weilin Zhao , Xu Han , Chaojun Xiao , Yingfa Chen , Yuxuan Li , Zhiyuan Liu , Maosong Sun

Generating realistic videos with diffusion transformers demands significant computation, with attention layers the central bottleneck; even producing a short clip requires running a transformer over a very long sequence of embeddings, e.g.,…

Computer Vision and Pattern Recognition · Computer Science 2025-09-23 Sankeerth Durvasula , Kavya Sreedhar , Zain Moustafa , Suraj Kothawade , Ashish Gondimalla , Suvinay Subramanian , Narges Shahidi , Nandita Vijaykumar

Robustifying convolutional neural networks (CNNs) against adversarial attacks remains challenging and often requires resource-intensive countermeasures. We explore the use of sparse mixture-of-experts (MoE) layers to improve robustness by…

Computer Vision and Pattern Recognition · Computer Science 2025-09-08 Svetlana Pavlitska , Haixi Fan , Konstantin Ditschuneit , J. Marius Zöllner

Mixture-of-Experts (MoE) architectures achieve scalable capacity through sparse routing, yet the geometric structure of expert specialization remains poorly understood. We introduce a unified Jacobian-PCA-Grassmann framework for analyzing…

Machine Learning · Computer Science 2026-05-19 Feilong Liu

Sparsity helps reduce the computational complexity of deep neural networks by skipping zeros. Taking advantage of sparsity is listed as a high priority in next generation DNN accelerators such as TPU. The structure of sparsity, i.e., the…

Machine Learning · Computer Science 2017-06-06 Huizi Mao , Song Han , Jeff Pool , Wenshuo Li , Xingyu Liu , Yu Wang , William J. Dally

The transformer architecture has prevailed in various deep learning settings due to its exceptional capabilities to select and compose structural information. Motivated by these capabilities, Sanford et al. proposed the sparse token…

Machine Learning · Statistics 2024-06-12 Zixuan Wang , Stanley Wei , Daniel Hsu , Jason D. Lee

Mixture-of-Experts (MoE) architectures have become the dominant choice for scaling Large Language Models (LLMs), activating only a subset of parameters per token. While MoE architectures are primarily adopted for computational efficiency,…

Computation and Language · Computer Science 2026-05-19 Jeremy Herbst , Stefan Wermter , Jae Hee Lee

Existing approaches to increasing the effective depth of Transformers predominantly rely on parameter reuse, extending computation through recursive execution. Under this paradigm, the network structure remains static along the training…

Computation and Language · Computer Science 2026-04-17 Yao Chen , Yilong Chen , Yinqi Yang , Junyuan Shang , Zhenyu Zhang , Zefeng Zhang , Shuaiyi Nie , Shuohuan Wang , Yu Sun , Hua Wu , HaiFeng Wang , Tingwen Liu

Traditional multi-task learning (MTL) methods use dense networks that use the same set of shared weights across several different tasks. This often creates interference where two or more tasks compete to pull model parameters in different…

Neural networks transform data through learned representations whose geometry affects separation, contraction, and generalization. Recent work studies this geometry using discrete curvature on neighborhood graphs, suggesting Ricci-flow-like…

Machine Learning · Computer Science 2026-05-05 Kanishka Reddy

Gradient descent on overparameterized neural networks typically operates at the Edge of Stability (EoS), where the largest Hessian eigenvalue hovers around a step-size-dependent threshold. We study how sparse connectivity changes…

Machine Learning · Statistics 2026-05-08 Tongtong Liang , Esha Singh , Rahul Parhi , Alexander Cloninger , Yu-Xiang Wang

In the past few years, neural networks have evolved from simple Feedforward Neural Networks to more complex neural networks, such as Convolutional Neural Networks and Recurrent Neural Networks. Where CNNs are a perfect fit for tasks where…

Machine Learning · Computer Science 2024-07-31 Harshil Darji

Recurrent neural networks (RNNs) are powerful dynamical models for data with complex temporal structure. However, training RNNs has traditionally proved challenging due to exploding or vanishing of gradients. RNN models such as LSTMs and…

Machine Learning · Computer Science 2020-06-17 Tankut Can , Kamesh Krishnamurthy , David J. Schwab

Federated learning (FL) enables collaborative training across organizations without sharing raw data, but it is hindered by statistical heterogeneity (non-i.i.d.\ client data) and by instability of naive weight averaging under client drift.…

This study delves into the application of graph neural networks in the realm of traffic forecasting, a crucial facet of intelligent transportation systems. Accurate traffic predictions are vital for functions like trip planning, traffic…

Machine Learning · Computer Science 2023-10-30 Razib Hayat Khan , Jonayet Miah , S M Yasir Arafat , M M Mahbubul Syeed , Duc M Ca

Latency and energy consumption are key metrics in the performance of deep neural network (DNN) accelerators. A significant factor contributing to latency and energy is data transfers. One method to reduce transfers or data is reusing data…

Hardware Architecture · Computer Science 2024-10-15 Michael Gilbert , Yannan Nellie Wu , Joel S. Emer , Vivienne Sze