AI & ML interests

Create NLP models and datasets applied to French, to very long sequences and the combination of the two ;)



Posts 3

view post
I stopped procrastinating and finally took the time to write the second article of my series of blog posts on SSM: https://huggingface.co/blog/lbourdois/ssm-2022.
In this blog post, I review the history of SSM models released in 2022, with over 14 models discussed in a synthetic format.
They are separated into two parts: "theoretical" (DSS, S4D, GSS, Mega, S5, etc.) and "applications" (Sashimi, ViS4mer, CCNN, etc.).

To understand everything, it's best to have read the introduction to S4 to SSM blog post first: https://huggingface.co/blog/lbourdois/get-on-the-ssm-train.
All the articles in the series are listed in this space: lbourdois/SSM_blog_posts

Wishing you a good reading :)
view post
The most widely used French NER models on HF ( Jean-Baptiste/camembert-ner and cmarkea/distilcamembert-base-ner) are trained on a single dataset (WikiNER) which on the one hand contains leaks and therefore distorts the true results of these models, and on the other hand overspecializes them in a particular domain (= texts from Wikipedia). They are also only available in a base version (110M parameters).

That's why I've trained new NER models in French both on more data (x3), as well as in base and large versions (336M). They are available in 3 entities (PER, ORG, LOC) or 4 entities (PER, ORG, LOC, MISC):
- CATIE-AQ/NERmembert-base-4entities
- CATIE-AQ/NERmembert-large-4entities
- CATIE-AQ/NERmembert-base-3entities
- CATIE-AQ/NERmembert-large-3entities

Datasets without leaks are also available:
- CATIE-AQ/frenchNER_4entities
- CATIE-AQ/frenchNER_3entities


None public yet