David Dale

cointegrated

AI & ML interests

Research engineer at FAIR, Meta. Some pet projects on NLP for under-resourced languages. Interests: Machine translation, Chatbots, applied NLU, controllable text generation (in particular, text style transfer), miniature models.

Recent Activity

new activity about 11 hours ago

openlanguagedata/flores_plus:(Beginner) Issues using the described method to load FLORES+ dataset

new activity about 22 hours ago

openlanguagedata/flores_plus:machine

new activity about 22 hours ago

openlanguagedata/flores_plus:Misalignments in the Aranese subset (aran1260)

View all activity

Organizations

cointegrated's activity

New activity in openlanguagedata/flores_plus about 11 hours ago

(Beginner) Issues using the described method to load FLORES+ dataset

#15 opened 1 day ago by

jaecbc

New activity in openlanguagedata/flores_plus about 22 hours ago

machine

#8 opened about 2 months ago by

maryamelboraie

Misalignments in the Aranese subset (aran1260)

#11 opened about 1 month ago by

OrianeN

Fix misalignments in the Aranese subset (aran1260)

#13 opened 14 days ago by

agaliano

New activity in openlanguagedata/flores_plus about 23 hours ago

Split dataset in subsets per language

#5 opened 3 months ago by

thomas-ferraz

New activity in openlanguagedata/flores_plus 19 days ago

[DRAFT] Fix orthography in the Russian dev set

#4 opened 3 months ago by

cointegrated

Fix encoding at chv devtest

#9 opened about 2 months ago by

alexantonov

New activity in slone/nllb-rus-tyv-v1 about 1 month ago

Adding `safetensors` variant of this model

#1 opened about 1 month ago by

SFconvertbot

New activity in cointegrated/LaBSE-en-ru about 1 month ago

Warn Some weights of the model checkpoint at cointegrated/LaBSE-en-ru were not used when initializing BertModel:

#4 opened 6 months ago by

alashkov83

New activity in slone/LaBSE-shallow-distilled-bak about 2 months ago

Adding `safetensors` variant of this model

#1 opened about 2 months ago by

SFconvertbot

New activity in cointegrated/SONAR_200_text_encoder about 2 months ago

can you please do the same for decoder

#2 opened 3 months ago by

damerajee

New activity in slone/finugorbib about 2 months ago

[bot] Conversion to Parquet

#1 opened about 2 months ago by

parquet-converter

New activity in openlanguagedata/flores_plus about 2 months ago

Added Dargwa dev set to flores_plus

#3 opened 4 months ago by

Murtazali

New activity in openlanguagedata/flores_plus 3 months ago

Add data integrity tests

#7 opened 3 months ago by

cointegrated

Two sentences in the dev set (one Lombard and one Tamasheq-Tifinagh) seem to be missing

#6 opened 3 months ago by

cointegrated

New activity in cointegrated/SONAR_200_text_encoder 5 months ago

Adding `safetensors` variant of this model

#1 opened 5 months ago by

SFconvertbot

New activity in slone/myv-rus-2022-quality-annotated 5 months ago

[bot] Conversion to Parquet

#1 opened 5 months ago by

parquet-converter

New activity in UNESCO/nllb 6 months ago

Optimize the preprocessing and generation

#11 opened 6 months ago by

cointegrated

New activity in cointegrated/SONAR-encoder-distilled-v0 10 months ago

More Details about the model

#1 opened 11 months ago by

sanjay73

New activity in cointegrated/rubert-tiny-toxicity 10 months ago

Training script?

#2 opened 10 months ago by

vdmbrsv