paper-with-me

Papers

Multi-Head LatentMoE and Head Parallel: Communication-Efficient and Deterministic MoE Parallelism

2026-02-04 · Chenwei Cui, Rockwell Jackson, Benjamin Joseph Herrera, Ana María Tárano, Hannah Kerner arxiv

Large language models have transformed many applications but remain expensive to train. Sparse Mixture of Experts (MoE) addresses this through conditional computation, with Expert Parallel (EP) as the standard distributed training method. However, EP has three limitations: communication cost grows linearly with the number of activated experts $k$, load imbalance affects latency and memory usage, and data-dependent communication requires metadata exchange. We propose Multi-Head LatentMoE and Head Parallel (HP), a new architecture and parallelism achieving $O(1)$ communication cost regardless of $k$, completely balanced traffic, and deterministic communication, all while remaining compatible with EP. To accelerate Multi-Head LatentMoE, we propose IO-aware routing and expert computation. Compared to MoE with EP, Multi-Head LatentMoE with HP trains up to $1.61\times$ faster while having identical performance. With doubled granularity, it achieves higher overall performance while still being $1.11\times$ faster. Our method makes multi-billion-parameter foundation model research more accessible.

📄 PDF Abstract BibTeX arXiv:2602.04870

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

ZeCO: Zero Communication Overhead Sequence Parallelism for Linear Attention

2025-07-01 · Yuhong Chou, Zehao Liu, Ruijie Zhu, Xinyi Wan 외 arxiv

Linear attention mechanisms deliver significant advantages for Large Language Models (LLMs) by providing linear computational complexity, enabling efficient processing of ultra-long sequences (e.g., 1M context). However,…

TACO: Efficient Communication Compression of Intermediate Tensors for Scalable Tensor-Parallel LLM Training

2026-04-27 · Man Liu, Xingchen Liu, Xingjian Tian, Bing Lu 외 arxiv

Handling communication overhead in large-scale tensor-parallel training remains a critical challenge due to the dense, near-zero distributions of intermediate tensors, which exacerbate errors under frequent communication…

EventGraD: Event-Triggered Communication in Parallel Machine Learning

2021-03-12 · Soumyadip Ghosh, Bernardo Aquino, Vijay Gupta

Communication in parallel systems imposes significant overhead which often turns out to be a bottleneck in parallel machine learning. To relieve some of this overhead, in this paper, we present EventGraD - an algorithm w…

BIG-bench Machine Learning

LCP: A Low-Communication Parallelization Method for Fast Neural Network Inference in Image Recognition

2020-03-13 · Ramyad Hadidi, Bahar Asgari, Jiashen Cao, Younmin Bae 외

Deep neural networks (DNNs) have inspired new studies in myriad edge applications with robots, autonomous agents, and Internet-of-things (IoT) devices. However, performing inference of DNNs in the edge is still a severe …

Quantization

Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training

2026-04-03 · Cunyang Wei, Siddharth Singh, Aishwarya Sarkar, Daniel Nichols 외 arxiv

Graph neural networks (GNNs) are widely used for learning on graph datasets derived from various real-world scenarios. Learning from extremely large graphs requires distributed training, and mini-batching with sampling i…