No Overfitting Lab

We are a group of independent researchers. We are working on the intersection of Natural Language Processing and Machine Learning and are focusing on:

  • Interpretability of big language models,
  • Evaluation of NLP applications,
  • Detection of mental health issues from text,
  • Few-shot learning.

News:

  • [August 2021] We have conducted research on whether BERT models are able to successfully learn depression specific language markers and symptoms from text. (Read more)
  • [July 2021] Our researchers are excited to be involved in the collaborative BigScience project (Read more)
  • [July 2021] Our task 'Estimating Risk of Suicide' was accepted to the BIG-Bench collaborative benchmark (Read more)