MAPA-Project

CEF project developing #anonymisation solutions for Public Administrations

[Pangeanic @Pangeanic]

Multilingual AI rarely fails everywhere at once. It fails locally.

Our new article explains language coverage, tokenization, annotation, evaluation and Local Quality Collapse.

https://t.co/YvjyQQyOul

#MultilingualAI #AIData

[Pangeanic @Pangeanic]

After the #Spanish government announcement labelling Palantir “the most dangerous company in the world” it makes sense to read our previous analysis on Captive AI https://t.co/W3YCOwOv84

[valgrai @fvalgrai]

🎥 Ya puedes ver la Company Talk de Manuel Herranz (@Pangeanic) en el #ValgrAI #VSCF2026, donde compartió la visión de la compañía sobre el papel de las ontologías, los modelos pequeños y la IA soberana en el futuro del desarrollo tecnológico.

🔗 https://t.co/diBACn4JsH

#IA #AI

[MAPA-Project @ProjectMapa]

I bet you didn’t know this: the website with the most #languages #translation https://t.co/Cd7hockr4v

[pangeanic_español @pangeanic_es]

Tecnologías del lenguaje 2023: ¡Estamos en el "#HypeCycle" de Gartner 2023! 😱

👉 Conoce nuestras capacidades y el porqué de nuestra mención:

https://t.co/jo4ek0FDIY

[MAPA-Project @ProjectMapa]

If Apple Translate is to improve, it better consider these parallel corpora tips https://t.co/Qpzf2har2I

[MAPA-Project @ProjectMapa]

Interesting analysis Embracing the Enigma: High Quality Data for Better AI Systems - NLP Consultancy https://t.co/codj4n6CWD

[qrdl @QRDL]

@ylecun @random_walker @sayashk we need more folks discussing distributed AI versus hegemonic AI and its implications with respect to AI safety

[Cristina Gallach Figueras @cristinagallach]

Sweden is developing its own big language model | Computer Weekly! Excellent project!!
https://t.co/N9zsxShla2

[EAMT2024 @EAMT_2024]

With most attendees of #EAMT2023 arriving in Finland today, we have one last Essential Finnish lesson prepared for you: how to say 'no' in Finnish. Now, brace yourselves for a twist: https://t.co/NfDbHo0ZOj

[AK @_akhaliq]

Meta just released MusicGen, a simple and controllable model for music generation

MusicGen is a single stage auto-regressive Transformer model trained over a 32kHz EnCodec tokenizer with 4 codebooks sampled at 50 Hz. Unlike existing methods like MusicLM, MusicGen doesn't not require a self-supervised semantic representation, and it generates all 4 codebooks in one pass. By introducing a small delay between the codebooks, can predict them in parallel, thus having only 50 auto-regressive steps per second of audio

try out the @Gradio demo: https://t.co/1UvsU6UTzS
Models on @huggingface: https://t.co/7uTFxSdhQa
github: https://t.co/d23lrVLhlE

[Yann LeCun @ylecun]

The evidence is accumulating.

[NTEU @NTEU3]

A resource for parallel data for #machinetranslation https://t.co/XxtxQENajS

[MAPA-Project @ProjectMapa]

We found a nice text data annotation service!! https://t.co/UzpSbbz26l

[Anna Kotarska @anna_kotarska]

6.3 million de-identified records
in NIH’s database
1 of the largest collections of covid patient records in the world
#anonymisation #deidentification #EHR
@CentrumeZdrowia @Centrum_MD @GovTechPL @twjedrzejczyk
i europ. rozwiązanie dla sektora publ. @ProjectMapa & @Pangeanic

[Pangeanic @Pangeanic]

Data anonymization refers to the process of de-identifying personal information from text. A type of information sanitization used to protect privacy. Learn more here: #Technology #TranslationTechnology #Translation https://t.co/AQjjQl977O

[Manuel Herranz @manuelhrrnz]

One of the most visionary articles I’ve read in a long time about #machinetranslation. A true vision for context aware, deep adaptive and multimodal MT based on multi technology NLP at its best, working towards real #AI
@ArleLommel https://t.co/6rKzbSg6lG #MTSummit2021

[Pangeanic @Pangeanic]

We are proud to announce that @CSA_Research has released its annual list of the 100 largest #LSP in the world and Pangeanic is ranked 8th out of the top 20 providers in Southern Europe.

#THANKS to our team for their tireless effort to improve day by day!

[MAPA-Project @ProjectMapa]

Glad to be mentioned by the security research rapid response to Covid-19 https://t.co/RLxC6EvM1d

[MAPA-Project @ProjectMapa]

@MarietjeSchaake No Marie Curie?