paper-with-me

Papers

Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models

2025-01-21 · Samira Abnar, Harshay Shah, Dan Busbridge, Alaaeldin Mohamed Elnouby Ali, Josh Susskind, Vimal Thilak

Scaling the capacity of language models has consistently proven to be a reliable approach for improving performance and unlocking new capabilities. Capacity can be primarily defined by two dimensions: the number of model parameters and the compute per example. While scaling typically involves increasing both, the precise interplay between these factors and their combined contribution to overall capacity remains not fully understood. We explore this relationship in the context of sparse Mixture-of-Experts (MoEs), which allow scaling the number of parameters without proportionally increasing the FLOPs per example. We investigate how varying the sparsity level, i.e., the fraction of inactive parameters, impacts model's performance during pretraining and downstream few-shot evaluation. We find that under different constraints (e.g., parameter size and total training compute), there is an optimal level of sparsity that improves both training efficiency and model performance. These results provide a better understanding of the impact of sparsity in scaling laws for MoEs and complement existing works in this area, offering insights for designing more efficient architectures.

📄 PDF Abstract BibTeX arXiv:2501.12370

Code (0)

등록된 구현이 없습니다.

Tasks

Mixture-of-Experts

Similar Papers 제목 키워드 기반

Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks

2025-08-26 · Taishi Nakamura, Satoki Ishikawa, Masaki Kawamura, Takumi Okamoto 외 arxiv

Empirical scaling laws have driven the evolution of large language models (LLMs), yet their coefficients shift whenever the model architecture or data pipeline changes. Mixture-of-Experts (MoE) models, now standard in st…

Reinforcement Learning

Generalizing Scaling Laws for Dense and Sparse Large Language Models

2025-08-08 · Md Arafat Hossain, Xingfu Wu, Valerie Taylor, Ali Jannesari arxiv

Despite recent advancements of large language models (LLMs), optimally predicting the model size for LLM pretraining or allocating optimal resources still remains a challenge. Several efforts have addressed the challenge…

Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies

2024-07-18 · Chaofan Tao, Qian Liu, Longxu Dou, Niklas Muennighoff 외

Research on scaling large language models (LLMs) has primarily focused on model parameters and training data size, overlooking the role of vocabulary size. We investigate how vocabulary size impacts LLM scaling laws by t…

ARC

Holistic Scaling Laws for Optimal Mixture-of-Experts Architecture Optimization

2026-03-23 · Weilin Wan, Jingtao Han, Weizhong Zhang, Cheng Jin arxiv

Scaling laws for Large Language Models govern macroscopic resource allocation, yet translating them into precise Mixture-of-Experts (MoE) architectural configurations remains an open problem due to the combinatorially va…

Scaling Laws For Diffusion Transformers

2024-10-10 · Zhengyang Liang, Hao He, Ceyuan Yang, Bo Dai

Diffusion transformers (DiT) have already achieved appealing synthesis and scaling properties in content recreation, e.g., image and video generation. However, scaling laws of DiT are less explored, which usually offer p…

Image GenerationText to Image GenerationText-to-Image GenerationVideo Generation