OLDI and friends
Collection
This collection groups the datasets that have been featured as part of WMT’s Open Language Data Initiative shared task. • 5 items • Updated • 5
Omar Kamali PRO
omarkamali
AI & ML interests
NLP & LLMs for low resource languages.
Recent Activity
liked a dataset about 7 hours ago
omneity-labs/ipa-dict repliedto their post 4 days ago
I just might have cracked tokenizer-free LLMs. No vocab, no softmax.
I'm training a 22M params LLM rn to test this "thing" and it's able to formulate coherent sentences 🤯
Bear in mind, this is a completely new, tokenizer-free LLM architecture with built-in language universality.
Check the explainer video to understand what's happening. Feedback welcome on this approach!
liked a Space 5 days ago
omneity-labs/lid-benchmarkOrganizations
upvoted a collection 7 days ago
upvoted a collection 8 months ago
upvoted an article over 1 year ago
Article
Finding Moroccan Arabic (Darija) in Fineweb 2
•
23