English
Related papers

Related papers: Parameters vs FLOPs: Scaling Laws for Optimal Spar…

200 papers

Mixture-of-Experts (MoE) architectures enable conditional computation by routing inputs to multiple expert subnetworks and are often motivated as a mechanism for scaling large language models. In this project, we instead study MoE behavior…

Machine Learning · Computer Science 2026-01-22 Adam Rokah , Daniel Veress , Caleb Caulk , Sourav Sharan

Mixture-of-Experts (MoE) enables efficient scaling of large language models (LLMs) with sparsely activated experts during inference. To effectively deploy large MoE models on memory-constrained devices, many systems introduce *expert…

Machine Learning · Computer Science 2026-03-03 Jingcong Liang , Siyuan Wang , Miren Tian , Yitong Li , Duyu Tang , Zhongyu Wei

Sparsely-gated Mixture of Experts networks (MoEs) have demonstrated excellent scalability in Natural Language Processing. In Computer Vision, however, almost all performant networks are "dense", that is, every input is processed by every…

Computer Vision and Pattern Recognition · Computer Science 2021-06-14 Carlos Riquelme , Joan Puigcerver , Basil Mustafa , Maxim Neumann , Rodolphe Jenatton , André Susano Pinto , Daniel Keysers , Neil Houlsby

Mixture-of-Experts (MoE) models enable efficient scaling of large language models (LLMs) by activating only a subset of experts per input. However, we observe that the commonly used auxiliary load balancing loss often leads to expert…

Computation and Language · Computer Science 2026-01-27 Hongcan Guo , Haolang Lu , Guoshun Nan , Bolun Chu , Jialin Zhuang , Yuan Yang , Wenhao Che , Xinye Cao , Sicong Leng , Qimei Cui , Xudong Jiang

Large language models are typically deployed as monolithic systems, requiring the full model even when applications need only a narrow subset of capabilities, e.g., code, math, or domain-specific knowledge. Mixture-of-Experts (MoEs)…

Computation and Language · Computer Science 2026-05-12 Ryan Wang , Akshita Bhagia , Sewon Min

We propose a novel scaling law for general-purpose decoder-only language models (LMs) trained on multilingual data, tackling the problem of balancing languages during multilingual pretraining. A primary challenge in studying multilingual…

Computation and Language · Computer Science 2024-12-05 Yifei He , Alon Benhaim , Barun Patra , Praneetha Vaddamanu , Sanchit Ahuja , Parul Chopra , Vishrav Chaudhary , Han Zhao , Xia Song

Improvements in language model capabilities are often attributed to increasing model size or training data, but in some cases smaller models trained on curated data or with different architectural decisions can outperform larger ones…

We develop a theory of generalization and scaling for Mixture-of-Experts (MoE) Transformers that cleanly separates \emph{active} per-input capacity from routing combinatorics. By conditioning on fixed routing patterns and union-bounding…

Machine Learning · Computer Science 2026-04-13 Mansour Zoubeirou a Mayaki

Sparse Mixture of Expert (MoE) models are popular foundational architectures at large scale, however, under-explored at smaller sizes. Here, we show how to enable Compact Sparse Mixture of Experts (CoSMoEs) for on-device inference.…

Machine Learning · Computer Science 2025-03-04 Patrick Huber , Akshat Shrivastava , Ernie Chang , Chinnadhurai Sankar , Ahmed Aly , Adithya Sagar

Despite their remarkable achievement, gigantic transformers encounter significant drawbacks, including exorbitant computational and memory footprints during training, as well as severe collapse evidenced by a high degree of parameter…

Machine Learning · Computer Science 2023-03-06 Tianlong Chen , Zhenyu Zhang , Ajay Jaiswal , Shiwei Liu , Zhangyang Wang

Large language models (LLMs) have garnered unprecedented advancements across diverse fields, ranging from natural language processing to computer vision and beyond. The prowess of LLMs is underpinned by their substantial model size,…

Machine Learning · Computer Science 2025-04-10 Weilin Cai , Juyong Jiang , Fan Wang , Jing Tang , Sunghun Kim , Jiayi Huang

Sparse Mixture of Experts (SMoE) enables efficient training of large language models by routing input tokens to a select number of experts. However, training SMoE remains challenging due to the issue of representation collapse. Recent…

Computation and Language · Computer Science 2025-04-01 Giang Do , Hung Le , Truyen Tran

Multilingual models are often particularly dependent on scaling to generalize to a growing number of languages. Compression techniques are widely relied upon to reconcile the growth in model size with real world resource constraints, but…

Computation and Language · Computer Science 2022-11-29 Kelechi Ogueji , Orevaoghene Ahia , Gbemileke Onilude , Sebastian Gehrmann , Sara Hooker , Julia Kreutzer

Mixture-of-Experts (MoE), a conditional computation architecture, achieved promising performance by scaling local module (i.e. feed-forward network) of transformer. However, scaling the cross-token module (i.e. self-attention) is…

Machine Learning · Computer Science 2022-01-17 Yuxuan Lou , Fuzhao Xue , Zangwei Zheng , Yang You

Sparse Mixture-of-Experts (MoE) architectures effectively scale model capacity by activating only a subset of experts for each input token. However, the standard Top-k routing strategy imposes a uniform sparsity pattern that ignores the…

Artificial Intelligence · Computer Science 2025-12-17 Can Jin , Hongwu Peng , Mingcan Xiang , Qixin Zhang , Xiangchi Yuan , Amit Hasan , Ohiremen Dibua , Yifan Gong , Yan Kang , Dimitris N. Metaxas

The Mixture of Experts (MoE) is a widely known neural architecture where an ensemble of specialized sub-models optimizes overall performance with a constant computational cost. However, conventional MoEs pose challenges at scale due to the…

Computation and Language · Computer Science 2023-09-12 Ted Zadouri , Ahmet Üstün , Arash Ahmadian , Beyza Ermiş , Acyr Locatelli , Sara Hooker

Many generation tasks follow a one-to-many mapping relationship: each input could be associated with multiple outputs. Existing methods like Conditional Variational AutoEncoder(CVAE) employ a latent variable to model this one-to-many…

Computation and Language · Computer Science 2021-08-19 Shaobo Cui , Xintong Bao , Xuming Lin , Zhongzhou Zhao , Ji Zhang , Wei Zhou , Haiqing Chen

Large Language Models (LLMs) have achieved remarkable success across many applications, with Mixture of Experts (MoE) models demonstrating great potential. Compared to traditional dense models, MoEs achieve better performance with less…

Machine Learning · Computer Science 2026-02-17 Zongle Huang , Lei Zhu , Zongyuan Zhan , Ting Hu , Weikai Mao , Xianzhi Yu , Yongpan Liu , Tianyu Zhang

Scaling large language models has driven remarkable advancements across various domains, yet the continual increase in model size presents significant challenges for real-world deployment. The Mixture of Experts (MoE) architecture offers a…

Machine Learning · Computer Science 2025-03-18 Shwai He , Daize Dong , Liang Ding , Ang Li

The increasing prevalence of sparse Mixture-of-Experts (MoE) architectures in large language models raises important questions regarding their reliability under stochastic decoding. While conditional computation enables substantial gains in…

Machine Learning · Computer Science 2026-01-06 Kabir Grover