Strange result that begs for future research: much bigger does not mean better? Or does it mean something else?

Last two weeks we participated in BiomedicalMT challenge for wmt22, and our MT model leads

Our fine-tuned model has taken the first place in Clinical Cases category of BiomedicalMT challenge for wmt22

We participated in BiomedicalMT challenge for wmt22, and our MT model leads in Clinical Cases

The path to signularity is not linear – it’ll be harder and harder

The inventor and visionary Ray Kurzweil, who coined the very term “singularity”, based his predictions on observation that today’s world is not linear.

Tuning Meta AI megatransformer ain’t so easy, and although the result is definitely better – but not without downside

On November 10, 2021 Meta AI Research published article “The first-ever multilingual model to win WMT, beating out bilingual models”.

  • 2022/07/09
  • -

AI and Quality Measurement are equivalent problems

In recent public discourse, claims were made that the need no longer exists to measure translation quality.

  • 2022/05/22
  • -

Paralela is featured in 337th Issue of Toolbox Journal

Jost Zetzsche has featured Paralela aligner in recent issue of Toolbox Journal. The article reads:

New, interesting research of Google Research and MIT on more effective human evaluation for machine translation, cites our work.

Google Research in cooperation with MIT Media Lab published new research “Toward More Effective Human Evaluation for Machine Translation”

  • 2021/09/02
  • -

Logrus Global releases Paralela, an AI-based aligner, as part of Logrus Global Localization Cloud

Sept 1, 2021 – In today’s data-driven world, high-quality bilingual data is a necessity for training better translation models as well as for capturing domain-specific knowledge