Moreno La Quatra

Our bilingual model for Speech-based Parkinson's works on different languages using specialized task heads 🎯 (... and other tweaks)

Great collaboration with Sabato Marco Siniscalchi and Juan Rafael Orozco-Arroyave!
👋 See you soon at Hyderabad for ICASSP 2025!

Moreno La Quatra, Valerio Mario Salerno, Yu Tsao, Sabato Marco Siniscalchi
FlanEC: Exploring Flan-T5 for Post-ASR Error Correction
https://arxiv.org/abs/2501.12979

Passionate about AI, music, and audio research?
Join our team at Stability AI as a research intern.

Application link: forms.gle/xmR2bup9AP1Y...

Fancy GenAI stuff like GPT 4 is too big, slow, private, and expensive for many jobs. Consider that the original GPT-1 was 117m params. Llama 3.1, by contrast, has up to 405 billion params! 😲

These models are slow, expensive, and *not yours to control*.

a man in a suit and tie is sitting in a chair with his hands in the air . ALT: a man in a suit and tie is sitting in a chair with his hands in the air .

Waiting for my bluesky feed to be full of #research updates, as it was for Twitter ~2Y ago, missing "following" or missing people on the platform?

academic poster presenting the results of the research project.

TweetyIta and ItaEval are a language model and evaluation benchmark for Italian tasks. What's more, they are 100% community-driven and born within RiTA (rita-nlp.org). @asantilli.bsky.social will present the poster on Dec 5, 16:30-17:30.

+ Pieter Delobelle, Moreno La Quatra, @bsavoldi.bsky.social

Thank you for sharing!