REVIEW 25 cited by
DiLoCo: Distributed Low-Communication Training of Language Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
DiLoCo: Distributed Low-Communication Training of Language Models
read the original abstract
Large language models (LLM) have become a critical component in many applications of machine learning. However, standard approaches to training LLM require a large number of tightly interconnected accelerators, with devices exchanging gradients and other intermediate states at each optimization step. While it is difficult to build and maintain a single computing cluster hosting many accelerators, it might be easier to find several computing clusters each hosting a smaller number of devices. In this work, we propose a distributed optimization algorithm, Distributed Low-Communication (DiLoCo), that enables training of language models on islands of devices that are poorly connected. The approach is a variant of federated averaging, where the number of inner steps is large, the inner optimizer is AdamW, and the outer optimizer is Nesterov momentum. On the widely used C4 dataset, we show that DiLoCo on 8 workers performs as well as fully synchronous optimization while communicating 500 times less. DiLoCo exhibits great robustness to the data distribution of each worker. It is also robust to resources becoming unavailable over time, and vice versa, it can seamlessly leverage resources that become available during training.
Forward citations
Cited by 25 Pith papers
-
What's in a Smoothness Constant? Tighter Rates for Local SGD with Bounded Second-order Heterogeneity
Local SGD provably improves over Mini-batch SGD under bounded second-order heterogeneity in the general convex setting, with nearly tight upper and lower bounds.
-
Cosine-Gated Adam-Decay: Drop-In Staleness-Aware Outer Optimization for Decoupled DiLoCo
CGAD is a staleness-aware Adam variant for DiLoCo that gates gradients with cosine and exponential decay, proves a convergence bound independent of maximum delay, and demonstrates stable pretraining of 25M to 7B param...
-
On the Surprising Effectiveness of a Single Global Merging in Decentralized Learning
A single global merge at the final step of decentralized SGD matches the convergence rate of parallel SGD while improving test accuracy under high data heterogeneity.
-
Federated Lightweight Fine-Tuning
A federated fine-tuning method transmits only 1,280 latent floats per round and reaches near-FedAvg accuracy by exploiting the exact averaging identity of affine mapping networks.
-
Byzantine Accountability Without Consensus: Strong Eventual Consistency for Non-Associative, Stochastic, Robust Aggregation
Any pure function of a product of CRDTs inherits Strong Eventual Consistency, so multi-Krum and similar discontinuous robust selectors can be made coordinator-free and accountable via OR-Set contributions plus grow-on...
-
Not Every Sync Is Safe: Calibrated DiLoCo Scheduling for Shared AI Infrastructure
Matched-random merge placement is the right control for DiLoCo scheduling claims; only calibrated overlap plus EWMA forecasting beats it in real vLLM sidecar replay (6.54%→5.09% SLO violations).
-
FoMoE: Breaking the Full-Replica Barrier with a Federation of MoEs
FoMoE partitions expert layers across workers in MoE LLMs, skips non-resident experts, and reports up to 1.42x lower communication than baselines plus 1.4x throughput gains while maintaining stable routing.
-
Unifying Local Communications and Local Updates for LLM Pretraining
GASLoC generalizes communication acceleration to the outer optimizer to enable gossip-based decentralized LLM pretraining that supports adaptive optimizers, local steps, and outperforms prior decentralized methods on ...
-
Learned Subspace Compression for Communication-Efficient Pipeline Parallelism
MAPL learns task-specific orthogonal compression subspaces per pipeline stage via manifold-constrained optimization and recovers signals with low-overhead anchors, yielding better compression-performance tradeoffs tha...
-
Local MixVR: Breaking the Communication-Sample Dependence in Distributed Learning
Local MixVR achieves communication complexity scaling only with number of workers M, independent of total samples N, and outperforms Minibatch Accelerated SGD when M is smaller than order N to the 1/4.
-
Optimistic Dual Averaging Unifies Modern Optimizers
SODA unifies several modern optimizers under optimistic dual averaging and supplies a 1/k decay wrapper that improves performance without weight decay tuning.
-
LoRDO: Distributed Low-Rank Optimization with Infrequent Communication
LoRDO combines global low-rank projections with full-rank quasi-hyperbolic momentum to let infrequent-synchronization distributed training match low-rank DDP at roughly 10x less communication.
-
Learning Decentralized LLM Collaboration with Multi-Agent Actor Critic
Multi-agent actor-critic methods with a centralized critic improve decentralized LLM collaboration over Monte Carlo baselines in long-horizon and sparse-reward settings.
-
Echelon: Auditable Aggregate-Only Language-Model Adaptation Across Privacy Boundaries
Echelon enables auditable aggregate-only adaptation of language models across privacy boundaries by training locally and sharing only boundary-level aggregates, achieving competitive performance in 1B LoRA experiments.
-
Outer-Momentum Restarting in High-Dimensional Two-Phase Optimization
Periodic outer-momentum restarts in two-phase optimizers exploit phase cancellation in a linearized NTK model to widen stable learning-rate and momentum ranges in language-model pretraining.
-
ScheduleFree+: Scaling Learning-Rate-Free & Schedule-Free Learning to Large Language Models
ScheduleFree+ scales schedule-free learning to LLMs with fixes for large batches and models, outperforming Warmup-Stable-Decay schedules by up to 31% at 1000 tokens per parameter.
-
Rennala MVR: Improved Time Complexity for Parallel Stochastic Optimization via Momentum-Based Variance Reduction
Rennala MVR improves time complexity over Rennala SGD for smooth nonconvex stochastic optimization in heterogeneous parallel systems under a mean-squared smoothness assumption.
-
Scalable Hyperparameter-Divergent Ensemble Training with Automatic Learning Rate Exploration for Large Models
HDET lets data-parallel replicas explore a spread of learning rates independently before averaging parameters, with an auto-LR controller driven by inter-replica loss differences to produce a self-adapting schedule wi...
-
Distributed Generative Inference of LLM at Internet Scales with Multi-Dimensional Communication Optimization
BloomBee is a distributed LLM inference system that achieves up to 1.76x higher throughput and 43.2% lower latency than prior decentralized systems by optimizing communication across multiple dimensions in low-bandwid...
-
Learning Decentralized LLM Collaboration with Multi-Agent Actor Critic
Centralized-critic actor-critic training (CoLLM-CC) improves sample efficiency and stability over Monte-Carlo multi-agent RL for training decentralized LLM collaboration.
-
Towards a future space-based, highly scalable AI infrastructure system design
Space-based AI compute is argued feasible via close-formation laser-linked satellites, radiation-survivable TPUs, and launch costs projected below $200/kg by the mid-2030s.
-
Controlled Periodic Synchronization for Efficient Data-Parallel Training
Periodic gradient+parameter synchronization with SlowMo beats DDP by 2.44 pp (K=4) on a WAN while cutting average wall-clock time by 13.8%, but only under a fixed LR=0.1 protocol.
-
ScaleAcross Explorer: Exploring Communication Optimization for Scale-Across AI Model Training
ScaleAcross Explorer jointly optimizes three design dimensions for scale-across training and reports up to 64.62% speedups over production baselines and 37.59% over prior art in testbed and simulation experiments.
-
Paris: A Decentralized Trained Open-Weight Diffusion Model
A text-to-image diffusion model trained with eight fully isolated experts plus a routing network matches centralized quality at much lower data/compute cost, according to the authors' measurements.
-
Quantum-Resilient Decentralized AI Economies: Proof-of-Useful-Work and Post-Quantum Security
Proposes Proof-of-Useful-Work for AI tasks in a token economy with post-quantum security, formalized via a closed-loop model and sufficient-stake condition.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.