Tweeted By @math_rachel

Lessons from NMT on 103 languages:
- Encoder reps cluster based on linguistic similarity
- Reps of source lang learned by the encoder depend on target language & viceversa
- Reps of high resource or similar langs are more robust when fine-tuning #WiML2019https://t.co/gwfakIUE2k pic.twitter.com/lLBb2cFyhj

— Rachel Thomas (@math_rachel) December 11, 2019