Accelerate Large Model Training using PyTorch Fully Sharded Data Parallel

Accelerate Large Model Training using DeepSpeed

Training Variable Long Sequences with Data-Centric Parallel

arXiv:2608. 07524v1 Announce Type: new Abstract: Training deep learning models on variable long sequences poses significant computational challenges.

DynamiQ: Accelerating Gradient Synchronization using Compressed Multi-hop All-reduce

arXiv:2602. 08923v2 Announce Type: replace Abstract: Multi-hop all-reduce is the de facto backbone of large model training.

Scaling Neural Network Verification with Tensor Parallelism and Fully Sharded Data Parallelism

arXiv:2606. 09377v1 Announce Type: cross Abstract: Formal neural network verification -- proving that a network satisfies safety properties for \emph{all} inputs in a specified domain -- is bounded in practice by GPU memory: standard implementations of bound-propagation algorithms (IBP, CROWN, $\alpha$-CROWN) require weight and relaxation-coefficient matrices to reside entirely on one accelerator.

Protocol Models: Scaling Decentralized Training with Communication-Efficient Model Parallelism

arXiv:2506. 01260v2 Announce Type: replace Abstract: Scaling models has led to significant advancements in deep learning, but training these models in decentralized settings remains challenging due to communication bottlenecks.

Subspace Networks: Scaling Decentralized Training with Communication-Efficient Model Parallelism

arXiv:2506. 01260v3 Announce Type: replace Abstract: Scaling models has led to significant advancements in deep learning, but training these models in decentralized settings remains challenging due to communication bottlenecks.

Demystifying Pipeline Parallelism: First Theory for PipeDream

arXiv:2606. 03498v1 Announce Type: new Abstract: Training modern machine learning models increasingly requires computation to be distributed across many accelerators.

ParVL: Parallel Scaling and Expandable Compute Allocation for Multimodal LLMs

Existing scaling strategies for Multimodal Large Language Models (MLLMs) typically expand either model parameters or sequential inference computation, incurring substantial memory or latency overhead. More importantly, most existing methods fail to alter the rigid, fixed computation allocation between the Vision Transformer and the Large Language Model components, limiting task-specific optimization.

How 🤗 Accelerate runs very large models thanks to PyTorch

SharQ: Bridging Activation Sparsity and FP4 Quantization for LLM Inference

arXiv:2606. 26587v1 Announce Type: cross Abstract: Low-bit floating-point formats and semi-structured sparsity are increasingly supported by modern accelerators, yet combining them for LLM activation compression remains challenging: activations contain input-dependent outliers that dominate block scales in FP4 quantization, and directly applying N:M sparsity masks discards moderate values, coupling sparsification loss with quantization error.

Scaling Neural Network Verification with Tensor Parallelism and Fully Sharded Data Parallelism

Formal neural network verification -- proving that a network satisfies safety properties for \emph{all} inputs in a specified domain -- is bounded in practice by GPU memory: standard implementations of bound-propagation algorithms (IBP, CROWN, $α$-CROWN) require weight and relaxation-coefficient matrices to reside entirely on one accelerator. We adapt two parallelism techniques originally developed for large-scale model training to the \texttt{auto\_LiRPA}\,/\,$α,β$-CROWN verification framework.

FlashCP: Load-Balanced Communication-Efficient Context Parallelism for LLM Training

arXiv:2606. 08476v1 Announce Type: cross Abstract: Context parallelism (CP) is essential for training large-scale, long-context language models, as it partitions sequences to reduce memory overhead.