PaddlePaddle: An Open-Source Deep Learning Platform from Industrial Practice

By Yanjun Ma, Dianhai Yu, Tian Wu, Haifeng Wang. 2019

353 citations recorded by the source corpus.

Subfields: Information Systems, Software, Human-Computer Interaction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Software

In Human-Computer Interaction

Citations in the corpus

RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering (2020, methodology): “Mao et al. (2020) used generation model to expand questions.”
Res2Net: A New Multi-Scale Backbone Architecture (2019, methodology): “Semi-supervised knowledge distillation solution [50] can also be applied to Res2Net, to achieve the 85.”
PAIR: Leveraging Passage-Centric Similarity Relation for Improving Dense Passage Retrieval (2021, methodology): “We conduct experiments with the deep learning framework PaddlePaddle (Ma et al., 2019) on up to eight NVIDIA Tesla V100 GPUs (with 32G RAM).”
Dense Text Retrieval Based on Pretrained Language Models: A Survey (2022, methodology): “For a fair comparison, we re-implement the DPR model with the library PaddlePaddele [246] and enhance it with the ERNIE model [176] as the base PLM14.”
AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results (2020, methodology): “For the experiment, the new model candidate search scheme using GP-NAS was implemented in PaddlePaddle [26] and the final-training of searched models were conducted using PyTorch.”
Interpretable deep learning: interpretation, interpretability, trustworthiness, and beyond (2021, methodology): “To simplify future researches and practical usages, we introduce several open-source libraries that implement popular interpretation algorithms based on mainstream deep learning frameworks, such as TF-Explainer4 based on Tensorflow [1], Captum5 based on PyTorch [125] and InterpretDL6 based on Paddle”
A Survey of Large Language Models (2023, neutral): “Besides the above libraries, existing deep learning frameworks (e.g., PyTorch [122], TensorFlow [123], MXNet [124], PaddlePaddle [125], MindSpore [99], and OneFlow [126]) have also provided support for parallelism algorithms, which are commonly used for training large-scale models.”
A Comprehensive Review of YOLO: From YOLOv1 to YOLOv8 and Beyond (2023, methodology): “authors used the PaddlePaddle [101] deep learning platform, hence its PP name.”
RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking (2021, methodology): “Natural Questions Test MRR@10 R@50 R@1000 R@5 R@20 R@100 BM25 (anserini) (Yang et al., 2017 ments with the deep learning framework Pad-dlePaddle (Ma et al., 2019) on up to 32 NVIDIA Tesla V100 GPUs (with 32G RAM).”
EdgeFlow: Achieving Practical Interactive Segmentation with Edge-Guided Flow (2021, neutral): “References [1] https://github.com/PaddlePaddle/Paddle.”
From distributed machine learning to federated learning: a survey (2021, methodology): “PaddleFL is an open-source federated learning framework based on PaddlePaddle [116], which is supported by Baidu.”
TurboTransformers: an efficient GPU serving system for transformer models (2020, methodology): “PaddlePaddle [17] used a similar method, which is all inspired by the caching device allocator implemented in the NVlab’s cub library [26].”
ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generation (2021, methodology): “PaddlePaddle [3] has also proposed 4D hybrid parallelism with more sophisticated combination techniques 6 .”
A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS (2023, methodology): “The authors used the PaddlePaddle [116] deep learning platform, hence its PP name.”