AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

GluonCV and GluonNLP: Deep Learning in Computer Vision and Natural Language Processing

Jian Guo, He He, Tong He, Leonard Lausen, Mu Li, Haibin Lin, Xingjian Shi, Chenguang Wang, Junyuan Xie, Sheng Zha, Aston Zhang, Hang Zhang and 4 more

JMLR 2020 · Vision-language and multimodal › Human and embodied modeling · approach: Human-like visual cognition in VLMs · 213 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use GluonCV training refinements (data augmentation, optimization) instead of Caffe referenceResNet-50 · ImageNettop-1 accuracy: 75.3 -> 79.2Improved
use GluonCV training refinements instead of Caffe referenceResNet-101 · ImageNettop-1 accuracy: 76.4 -> 80.5Improved
use GluonCV implementation instead of TensorFlow referenceMobileNet 1.0 · ImageNettop-1 accuracy: 70.9 -> 73.3Improved
use GluonCV implementation instead of DetectronFaster R-CNN · COCO detectionmAP: 39.6 -> 40.1Improved
use GluonCV implementation instead of DetectronMask R-CNN · COCO instance segmentationmask AP: 32.8 -> 33.1Improved
use GluonCV implementation instead of mmactionI3D ResNet-50 · Kinetics400top-1 accuracy: 72.9 -> 74.0Improved
use GluonNLP implementation instead of referencetextCNN · TRECaccuracy: 92.2 -> 92.8Improved
use GluonNLP implementation instead of Google BERTBERT-base · MRPCaccuracy: 86.7 -> 88.7Improved
use GluonNLP implementation instead of Google BERTBERT-base · MNLI-maccuracy: 84.4 -> 84.6Improved
use GluonNLP implementation instead of Google BERTBERT-base · SST-2accuracy: 92.7 -> 93.0Improved
use GluonNLP implementation instead of Google BERTBERT-base · SQuAD 1.1F1/EM: 88.5/80.8 -> 88.5/81.0No clear effect
use GluonNLP implementation instead of Google BERTBERT-large · SQuAD 1.1F1/EM: 90.9/84.1 -> 91.0/84.1No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy