Cover Image

Communication Efficient LLM Pre-training with SparseLoCo

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Sarfi, Amir, Thérien, Benjamin, Lidin, Joel, Belilovsky, Eugene
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Heterogeneous Low-Bandwidth Pre-Training of LLMs
by: Obeidi, Yazan, et al.
Published: (2026)

Covenant-72B: Pre-Training a 72B LLM with Trustless Peers Over-the-Internet
by: Lidin, Joel, et al.
Published: (2026)

Incentivizing Permissionless Distributed Learning of LLMs
by: Lidin, Joel, et al.
Published: (2025)

MuLoCo: Muon is a practical inner optimizer for DiLoCo
by: Thérien, Benjamin, et al.
Published: (2025)

Meta-learning Optimizers for Communication-Efficient Learning
by: Joseph, Charles-Étienne, et al.
Published: (2023)

Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training
by: Singh, Vaibhav, et al.
Published: (2025)

Simple and Scalable Strategies to Continually Pre-train Large Language Models
by: Ibrahim, Adam, et al.
Published: (2024)

$μ$LO: Compute-Efficient Meta-Generalization of Learned Optimizers
by: Thérien, Benjamin, et al.
Published: (2024)

Understanding and Exploiting Weight Update Sparsity for Communication-Efficient Distributed RL
by: Miahi, Erfan, et al.
Published: (2026)

PyLO: Towards Accessible Learned Optimizers in PyTorch
by: Janson, Paul, et al.
Published: (2025)

Continual Pre-training of MoEs: How robust is your router?
by: Thérien, Benjamin, et al.
Published: (2025)

Model Breadcrumbs: Scaling Multi-Task Model Merging with Sparse Masks
by: Davari, MohammadReza, et al.
Published: (2023)

Efficient Refusal Ablation in LLM through Optimal Transport
by: Nanfack, Geraldin, et al.
Published: (2026)

Stabilizing Native Low-Rank LLM Pretraining
by: Janson, Paul, et al.
Published: (2026)

Warming Up for Zeroth-Order Federated Pre-Training with Low Resource Clients
by: Legate, Gwen, et al.
Published: (2025)

FairDropout: Using Example-Tied Dropout to Enhance Generalization of Minority Groups
by: Nanfack, Geraldin, et al.
Published: (2025)

ACCO: Accumulate While You Communicate for Communication-Overlapped Sharded LLM Training
by: Nabli, Adel, et al.
Published: (2024)

MediSwift: Efficient Sparse Pre-trained Biomedical Language Models
by: Thangarasa, Vithursan, et al.
Published: (2024)

When Data Falls Short: Grokking Below the Critical Threshold
by: Singh, Vaibhav, et al.
Published: (2025)

Dual-Phase Continual Learning: Supervised Adaptation Meets Unsupervised Retention
by: Singh, Vaibhav, et al.
Published: (2024)

Celo2: Towards Learned Optimization Free Lunch
by: Moudgil, Abhinav, et al.
Published: (2026)

Celo: Training Versatile Learned Optimizers on a Compute Diet
by: Moudgil, Abhinav, et al.
Published: (2025)

WASH: Train your Ensemble with Communication-Efficient Weight Shuffling, then Average
by: Fournier, Louis, et al.
Published: (2024)

Not Only the Last-Layer Features for Spurious Correlations: All Layer Deep Feature Reweighting
by: Hameed, Humza Wajid, et al.
Published: (2024)

S-STE: Continuous Pruning Function for Efficient 2:4 Sparse Pre-training
by: Hu, Yuezhou, et al.
Published: (2024)

DragD3D: Realistic Mesh Editing with Rigidity Control Driven by 2D Diffusion Priors
by: Xie, Tianhao, et al.
Published: (2023)

SparseLoRA: Accelerating LLM Fine-Tuning with Contextual Sparsity
by: Khaki, Samir, et al.
Published: (2025)

Dense Backpropagation Improves Training for Sparse Mixture-of-Experts
by: Panda, Ashwinee, et al.
Published: (2025)

Rethinking Prompt Optimization: Reinforcement, Diversification, and Migration in Blackbox LLMs
by: Davari, MohammadReza, et al.
Published: (2025)

From Feature Visualization to Visual Circuits: Effect of Adversarial Model Manipulation
by: Nanfack, Geraldin, et al.
Published: (2024)

Less is More: Undertraining Experts Improves Model Upcycling
by: Horoi, Stefan, et al.
Published: (2025)

Adaptive Intelligent Secondary Control of Microgrids Using a Biologically-Inspired Reinforcement Learning
by: Jafari, Mohammad, et al.
Published: (2019)

LOST: Low-rank and Sparse Pre-training for Large Language Models
by: Li, Jiaxi, et al.
Published: (2025)

The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws
by: Jin, Tian, et al.
Published: (2025)

C-LoRA: Continual Low-Rank Adaptation for Pre-trained Models
by: Zhang, Xin, et al.
Published: (2025)

Visual Deformation Detection Using Soft Material Simulation for Pre-training of Condition Assessment Models
by: Sol, Joel, et al.
Published: (2024)

Path-independent Flow Matching for Multi-parameter Generative Dynamics
by: Téllez, Francisco, et al.
Published: (2026)

Model Parallelism With Subnetwork Data Parallelism
by: Singh, Vaibhav, et al.
Published: (2025)

Non-Uniform Parameter-Wise Model Merging
by: Camacho, Albert Manuel Orozco, et al.
Published: (2024)

Federated LoRA with Sparse Communication
by: Kuo, Kevin, et al.
Published: (2024)