paper-with-me

Papers

Communication-Efficient Language Model Training Scales Reliably and Robustly: Scaling Laws for DiLoCo

2025-03-12 · Zachary Charles, Gabriel Teston, Lucio Dery, Keith Rush, Nova Fallen, Zachary Garrett, Arthur Szlam, Arthur Douillard

As we scale to more massive machine learning models, the frequent synchronization demands inherent in data-parallel approaches create significant slowdowns, posing a critical challenge to further scaling. Recent work develops an approach (DiLoCo) that relaxes synchronization demands without compromising model quality. However, these works do not carefully analyze how DiLoCo's behavior changes with model size. In this work, we study the scaling law behavior of DiLoCo when training LLMs under a fixed compute budget. We focus on how algorithmic factors, including number of model replicas, hyperparameters, and token budget affect training in ways that can be accurately predicted via scaling laws. We find that DiLoCo scales both predictably and robustly with model size. When well-tuned, DiLoCo scales better than data-parallel training with model size, and can outperform data-parallel training even at small model sizes. Our results showcase a more general set of benefits of DiLoCo than previously documented, including increased optimal batch sizes, improved downstream generalization with scale, and improved evaluation loss for a fixed token budget.

📄 PDF Abstract BibTeX arXiv:2503.09799

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage Modelling

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
Focus 설명 없음

Similar Papers 제목 키워드 기반

Reformulating NLP tasks to Capture Longitudinal Manifestation of Language Disorders in People with Dementia

2023-10-15 · Dimitris Gkoumas, Matthew Purver, Maria Liakata

Dementia is associated with language disorders which impede communication. Here, we automatically learn linguistic disorder patterns by making use of a moderately-sized pre-trained language model and forcing it to focus …

Language ModelingLanguage Modelling

Efficient Adversarial Training in LLMs with Continuous Attacks

2024-05-24 · Sophie Xhonneux, Alessandro Sordoni, Stephan Günnemann, Gauthier Gidel 외

Large language models (LLMs) are vulnerable to adversarial attacks that can bypass their safety guardrails. In many domains, adversarial training has proven to be one of the most promising methods to reliably improve rob…

DAISY Filter Flow: A Generalized Discrete Approach to Dense Correspondences

2014-06-01 · CVPR 2014 6 · Hongsheng Yang, Wen-Yan Lin, Jiangbo Lu

Establishing dense correspondences reliably between a pair of images is an important vision task with many applications. Though significant advance has been made towards estimating dense stereo and optical flow fields fo…

Optical Flow Estimation

Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs

2026-02-19 · Luke J. Huang, Zhuoyang Zhang, Qinghao Hu, Shang Yang 외 arxiv

Asynchronous reinforcement learning has become increasingly central to scaling LLM post-training, delivering major throughput gains by decoupling rollout generation from policy updates. However, widely used policy-gradie…

Reinforcement Learning

A Scalable Communication Protocol for Networks of Large Language Models

2024-10-14 · Samuele Marro, Emanuele La Malfa, Jesse Wright, Guohao Li 외

Communication is a prerequisite for collaboration. When scaling networks of AI-powered agents, communication must be versatile, efficient, and portable. These requisites, which we refer to as the Agent Communication Tril…