AutoLoCo: Communication Efficient Distributed LLM Training via Adaptive Synchronization
cs.LG, cs.AI
Submitted: 2026-09-29
Updated: 2026-09-29
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Enhancing Chat Language Models by Scaling High-quality Instructional Conversations
- DiLoCo: Distributed Low-Communication Training of Language Models
- Streaming DiLoCo with overlapping communication: Towards a Distributed Free Lunch
- The Llama 3 Herd of Models
- OpenDiLoCo: An Open-Source Framework for Globally Distributed Low-Communication Training
- Adaptive Periodic Averaging: A Practical Approach to Reducing Communication in Distributed Learning
- SNOO: Step-K Nesterov Outer Optimizer - The Surprising Effectiveness of Nesterov Momentum Applied to Pseudo-Gradients
- PyTorch Distributed: Experiences on Accelerating Data Parallel Training
- LLaMA: Open and Efficient Foundation Language Models
- Instruction-Following Evaluation for Large Language Models
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks