Computational Linguistics @ UZH

@cl-uzh.bsky.social

cl.uzh.ch

Congratulations to Iuliia Thorbecke (Nigmatulina), who just passed the viva of her PhD on using contextual information to improve automatic speech recognition, for example using radar flight information to better transcribe air traffic communication.

Iuliia Thorbecke wearing her Docotor's hat that was decorated by other PhD students of our department

🚨New Preprint! In multilingual models, the same meaning can take far more tokens in some languages, penalizing users of underrepresented languages with worse performance and higher API costs. Our Parity-aware BPE algorithm is a step toward addressing this issue: 🧵

Bild