{"list":"other","total":64,"rows":[{"id":"1706.03762","title":"Attention Is All You Need","created":"2017-06-12","cited_by_count":60361},{"id":"1810.04805","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","created":"2018-10-11","cited_by_count":40399},{"id":"2005.14165","title":"Language Models are Few-Shot Learners","created":"2020-05-28","cited_by_count":32419},{"id":"1912.01703","title":"PyTorch: An Imperative Style, High-Performance Deep Learning Library","created":"2019-12-03","cited_by_count":20305},{"id":"1910.10683","title":"Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer","created":"2019-10-23","cited_by_count":14599},{"id":"1905.11946","title":"EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks","created":"2019-05-28","cited_by_count":6442},{"id":"1911.11641","title":"PIQA: Reasoning about Physical Commonsense in Natural Language","created":"2019-11-26","cited_by_count":2883},{"id":"1909.08053","title":"Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism","created":"2019-09-17","cited_by_count":1868},{"id":"1706.02677","title":"Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour","created":"2017-06-08","cited_by_count":1492},{"id":"1910.02054","title":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","created":"2019-10-04","cited_by_count":1301},{"id":"1606.06031","title":"The LAMBADA dataset: Word prediction requiring a broad discourse context","created":"2016-06-20","cited_by_count":992},{"id":"1802.01548","title":"Regularized Evolution for Image Classifier Architecture Search","created":"2018-02-05","cited_by_count":909},{"id":"1704.04683","title":"RACE: Large-scale ReAding Comprehension Dataset From Examinations","created":"2017-04-15","cited_by_count":861},{"id":"1604.06174","title":"Training Deep Nets with Sublinear Memory Cost","created":"2016-04-21","cited_by_count":848},{"id":"1905.13319","title":"MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms","created":"2019-05-30","cited_by_count":662},{"id":"2104.04473","title":"Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM","created":"2021-04-09","cited_by_count":632},{"id":"1704.04760","title":"In-Datacenter Performance Analysis of a Tensor Processing Unit","created":"2017-04-16","cited_by_count":618},{"id":"1802.04434","title":"signSGD: Compressed Optimisation for Non-Convex Problems","created":"2018-02-13","cited_by_count":571},{"id":"1905.12616","title":"Defending Against Neural Fake News","created":"2019-05-29","cited_by_count":559},{"id":"2201.11990","title":"Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model","created":"2022-01-28","cited_by_count":559},{"id":"1712.01887","title":"Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training","created":"2017-12-05","cited_by_count":478},{"id":"2101.06840","title":"ZeRO-Offload: Democratizing Billion-Scale Model Training","created":"2021-01-18","cited_by_count":341},{"id":"1404.5997","title":"One weird trick for parallelizing convolutional neural networks","created":"2014-04-23","cited_by_count":264},{"id":"1705.07878","title":"TernGrad: Ternary Gradients to Reduce Communication in Distributed Deep Learning","created":"2017-05-22","cited_by_count":256},{"id":"2104.07857","title":"ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning","created":"2021-04-16","cited_by_count":256},{"id":"1909.01315","title":"Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks","created":"2019-09-03","cited_by_count":242},{"id":"1905.13727","title":"PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization","created":"2019-05-31","cited_by_count":238},{"id":"1806.02847","title":"A Simple Method for Commonsense Reasoning","created":"2018-06-07","cited_by_count":223},{"id":"1808.07217","title":"Don't Use Large Mini-Batches, Use Local SGD","created":"2018-08-22","cited_by_count":202},{"id":"1906.02367","title":"Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification, and Local Computations","created":"2019-06-06","cited_by_count":181},{"id":"1807.05358","title":"Beyond Data and Model Parallelism for Deep Neural Networks","created":"2018-07-14","cited_by_count":179},{"id":"2007.01045","title":"DAPPLE: A Pipelined Data Parallel Approach for Training Large Models","created":"2020-07-02","cited_by_count":101},{"id":"1412.6651","title":"Deep learning with Elastic Averaging SGD","created":"2014-12-20","cited_by_count":90},{"id":"2006.15704","title":"PyTorch Distributed: Experiences on Accelerating Data Parallel Training","created":"2020-06-28","cited_by_count":90},{"id":"2102.02888","title":"1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed","created":"2021-02-04","cited_by_count":75},{"id":"2107.06925","title":"Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines","created":"2021-07-14","cited_by_count":68},{"id":"2103.03936","title":"Pufferfish: Communication-efficient Models At No Extra Cost","created":"2021-03-05","cited_by_count":61},{"id":"1706.03292","title":"Poseidon: An Efficient Communication Architecture for Distributed Deep Learning on GPU Clusters","created":"2017-06-11","cited_by_count":57},{"id":"1910.05124","title":"PipeMare: Asynchronous Pipeline Parallel DNN Training","created":"2019-10-09","cited_by_count":55},{"id":"2010.00091","title":"Error Compensated Distributed SGD Can Be Accelerated","created":"2020-09-30","cited_by_count":45},{"id":"2010.12460","title":"Adaptive Gradient Quantization for Data-Parallel SGD","created":"2020-10-23","cited_by_count":45},{"id":"2203.10428","title":"PipeGCN: Efficient Full-Graph Training of Graph Convolutional Networks with Pipelined Feature Communication","created":"2022-03-20","cited_by_count":45},{"id":"2006.11077","title":"A Better Alternative to Error Feedback for Communication-Efficient Distributed Learning","created":"2020-06-19","cited_by_count":43},{"id":"2105.05720","title":"Breaking the Computation and Communication Abstraction Barrier in Distributed Machine Learning Workloads","created":"2021-05-12","cited_by_count":40},{"id":"1910.04940","title":"Blink: Fast and Generic Collectives for Distributed ML","created":"2019-10-11","cited_by_count":36},{"id":"1712.02679","title":"AdaComp : Adaptive Residual Gradient Compression for Data-Parallel Distributed Training","created":"2017-12-07","cited_by_count":33},{"id":"2101.10761","title":"An Efficient Statistical-based Gradient Compression Technique for Distributed Training Systems","created":"2021-01-26","cited_by_count":32},{"id":"1809.02839","title":"Efficient and Robust Parallel DNN Training through Model Parallelism on Multi-GPU Platform","created":"2018-09-08","cited_by_count":28},{"id":"1803.03288","title":"TicTac: Accelerating Distributed Deep Learning with Communication Scheduling","created":"2018-03-08","cited_by_count":27},{"id":"2103.00543","title":"On the Utility of Gradient Compression in Distributed Training Systems","created":"2021-02-28","cited_by_count":26},{"id":"2010.10458","title":"Towards Scalable Distributed Training of Deep Learning on Public Cloud Clusters","created":"2020-10-20","cited_by_count":23},{"id":"2201.07598","title":"Near-Optimal Sparse Allreduce for Distributed Deep Learning","created":"2022-01-19","cited_by_count":23},{"id":"1711.04325","title":"Extremely Large Minibatch SGD: Training ResNet-50 on ImageNet in 15 Minutes","created":"2017-11-12","cited_by_count":22},{"id":"2104.11125","title":"ScaleCom: Scalable Sparsified Gradient Compression for Communication-Efficient Distributed Training","created":"2021-04-21","cited_by_count":22},{"id":"1905.03960","title":"Priority-based Parameter Propagation for Distributed DNN Training","created":"2019-05-10","cited_by_count":21},{"id":"1805.10111","title":"Double Quantization for Communication-Efficient Distributed Optimization","created":"2018-05-25","cited_by_count":19},{"id":"1901.02067","title":"HyPar: Towards Hybrid Parallelism for Deep Learning Accelerator Array","created":"2019-01-07","cited_by_count":15},{"id":"2003.11666","title":"Pipelined Backpropagation at Scale: Training Large Models without Batches","created":"2020-03-25","cited_by_count":11},{"id":"1907.13257","title":"Optimizing Multi-GPU Parallelization Strategies for Deep Learning Training","created":"2019-07-30","cited_by_count":9},{"id":"2008.05823","title":"Step-Ahead Error Feedback for Distributed Training with Compressed Gradient","created":"2020-08-13","cited_by_count":9},{"id":"1602.06709","title":"Distributed Deep Learning Using Synchronous Stochastic Gradient Descent","created":"2016-02-22","cited_by_count":8},{"id":"1711.04291","title":"Scale out for large minibatch SGD: Residual network training on ImageNet-1K with improved accuracy and reduced time to train","created":"2017-11-12","cited_by_count":8},{"id":"2006.02924","title":"Scaling Distributed Training with Adaptive Summation","created":"2020-06-04","cited_by_count":4},{"id":"1805.04170","title":"Unifying Data, Model and Hybrid Parallelism in Deep Learning via Tensor Tiling","created":"2018-05-10","cited_by_count":3}],"next_cursor":null}