Mihaela Grigore
  • 👋About
  • 👩‍🏭Personal projects
    • Computer Vision | Deep Learning with Tensorflow & Keras (ResNet50, GPU training)
    • Computer Vision | Convolutional Neural Networks with PyTorch
    • Computer Vision | Facial Recognition with Keras, FaceNet, Inception, Siamese Networks
    • NLP | Topic modeling on tweets
    • NLP | Sentiment analysis of tweets: TextBlob, VADER and Flair
    • Time series | Exploration on Crypto price dataset
    • Data scraping | Social Media Scraping: Twitter Developer API for Academics
    • Data Scraping | Collecting historical tweets without Twitter API
  • ✍️Notes
    • Machine Learning in Production
      • Feature transforms
      • Feature selection
      • Data journey
    • NLP
      • Information Retrieval
    • Computer Vision
    • Time series
      • Stationarity
    • Data
      • Labeling
    • Python
      • ndarray slicing with index out of bounds
  • 📚Readings & other media
    • Computer Vision
      • Selection of research articles
    • NLP
      • Handwriting Text
      • Information Retrieval
      • Mono- / multilingual
      • Topic Modeling
      • Language Models
    • Time Series
    • Generative Adversarial Netoworks (GAN)
    • Python
      • Python basics
Powered by GitBook
On this page
  1. Readings & other media
  2. NLP

Mono- / multilingual

  • Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation, by Nils Reimers, Iryna Gurevych [ paper ] An easy and efficient method to extend existing sentence embedding models to new languages. This allows to create multilingual versions from previously monolingual models.

  • Multilingual Universal Sentence Encoder for Semantic Retrieval, by Yinfei Yang, Daniel Cer, Amin Ahmad, Mandy Guo, Jax Law, Noah Constant, Gustavo Hernandez Abrego, Steve Yuan, Chris Tar, Yun-Hsuan Sung, Brian Strope, Ray Kurzweil [ paper ] When the authors team is from Google and it comprises of 12 people, it's a sign of something worth reading. - They propose two pre-trained multilingual (16 languages) retrieval models based on Transformer and CNN model architectures. - They reach SOTA performance on the folowing tasks: semantic retrieval (SR), translation pair bitext retrieval (BR) and retrieval question answering (ReQA). - They make the models available on TensorFlow Hub. - They also wrote a short blog post about their findngs on Google Blog.

PreviousInformation RetrievalNextTopic Modeling

Last updated 2 years ago

📚