Language Models Could Learn Semantics - No Matter How You Define It

Language Models Could Learn Semantics - No Matter How You Define It

Santa Fe Institute via YouTube Direct link

Assumptions we need to prove this theorem

9 of 17

9 of 17

Assumptions we need to prove this theorem

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Language Models Could Learn Semantics - No Matter How You Define It

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 What I do
  3. 3 Learning meaning from form
  4. 4 Overview
  5. 5 An ideal language mode
  6. 6 Entailment semantics
  7. 7 Gricean speakers
  8. 8 Example
  9. 9 Assumptions we need to prove this theorem
  10. 10 Experiment in toy settings
  11. 11 Experiment: MNLI
  12. 12 Limitation 1: the no-redundancy assumption is too strong
  13. 13 How close can we get to the ideal language model in practice?
  14. 14 Interim discussion
  15. 15 Back to reference
  16. 16 Can language models refer
  17. 17 Conclusions

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.