mt@upc

Neural Machine Translation. Universitat Politècnica de Catalunya

[Javier Ferrando @javifer_96]

[1/4] Introducing “A Primer on the Inner Workings of Transformer-based Language Models”, a comprehensive survey on interpretability methods and the findings into the functioning of language models they have led to.

ArXiv: https://t.co/UBh2ZLmYkr

[mt@upc @mtupc1]

Microsoft apuesta por Barcelona como hub europeo con 100 nuevas contrataciones.

https://t.co/WlWTqe3uwa

[Belen Alastruey @b_alastruey]

Honored to share that I’ve been granted the ICTWoman of the year 2022 award for university students, by the Govt. of Catalonia @tic. This award recognizes women with remarkable careers, to foster girls vocations in ICT. Big thanks to everyone that has helped me get here!☺️

[Javier Ferrando @javifer_96]

What a day!!

2 papers accepted at #EMNLP2022! 🥳

Big thanks to all my collaborators. Special thanks to the amazing @geiongallego.

[Belen Alastruey @b_alastruey]

Happy to share “On the Locality of Attention in Direct Speech Translation”💬🔍 accepted at the ACL-SRW!

We use interpretability techniques to
propose an efficient architecture that
matches the baseline performance while reducing computational cost.

https://t.co/nTduu8pcAB

[Javier Ferrando @javifer_96]

Happy to share our work on interpretability of the Transformer. Our proposed method, ALTI (Aggregation of Layer-wise Token-to-token Interactions), accurately measures how contextual information is mixed across the Transformer. https://t.co/97cLypEH6e

[Bcn Analytics @bcnanalytics]

Update about our Datathon against racism. Save the dates: March 25th and April 1st. More details, here https://t.co/i6V9c5IvIG

[Christine Basta @Christine_Basta]

We are so delighted that our paper 'Interpreting Gender Bias in Neural Machine Translation: The Multilingual Architecture Matters' has been accepted at #AAAI2022.

[mt@upc @mtupc1]

And another post on Efficient Transformer for Direct Speech Translation by Belén Alastruey. https://t.co/W4NVYnhD5H

[mt@upc @mtupc1]

New mt@upc post about dysarthric speech synthesis by Marc Illa
https://t.co/ZgepgNEdjY

[mt@upc @mtupc1]

We are very pleased to announce that our paper “Attention Weights in Transformer NMT Fail Aligning Words Between Sequences but Largely Explain Model Predictions” by @javifer_96 and @costajussamarta has been accepted at #EMNLP2021 as a Findings paper

[mt@upc @mtupc1]

Interested in Lifelong Learning in Neural Machine Translation? Check out our last blog post by Lluís Guardia https://t.co/KGE3bPApGA

[mt@upc @mtupc1]

Do not miss our blogs about The Transformer, Neural Machine Translation, Multilingual MT, End-to-End Speech Translation, Unsupervised NMT, Factored NMT, Interpretability in NMT, Gender bias in NLP, Lifelong Learning and Knowledge Generation and Discovery. https://t.co/Fmgd5C5wbr

[mt@upc @mtupc1]

Hello World!
Happy to launch our Twitter account to keep you updated about our activities and publications. https://t.co/hBMUUyQlTh