Unsupervised Cross-lingual Representation Learning at Scale
This paper shows that pretraining multilingual language models at scale leads to significant performance gains for a wide range of cross-lingual transfer tasks. We train a Transformer-based masked language model on one hundred languages, using more than two terabytes of filtered CommonCrawl data. Our model, dubbed XLM-R, significantly outperforms multilingual BERT (mBERT) on a variety of cross-lingual benchmarks, including +13.8% average accuracy on XNLI, +12.3% average F1 score on MLQA, and +2.1% average F1 score on NER. XLM-R performs particularly well on low-resource languages, improving 11.8% in XNLI accuracy for Swahili and 9.2% for Urdu over the previous XLM model. We also present a detailed empirical evaluation of the key factors that are required to achieve these gains, including the trade-offs between (1) positive transfer and capacity dilution and (2) the performance of high and low resource languages at scale. Finally, we show, for the first time, the possibility of multilingual modeling without sacrificing per-language performance; XLM-Ris very competitive with strong monolingual models on the GLUE and XNLI benchmarks. We will make XLM-R code, data, and models publicly available.
NurtureToken New!

Token crowdsale for this paper ends in

Buy Nurture Tokens

Authors

Are you an author of this paper? Check the Twitter handle we have for you is correct.

Alexis Conneau (edit)
Kartikay Khandelwal (edit)
Naman Goyal (edit)
Vishrav Chaudhary (add twitter)
Guillaume Wenzek (add twitter)
Francisco Guzmán (edit)
Edouard Grave (edit)
Myle Ott (edit)
Luke Zettlemoyer (add twitter)
Veselin Stoyanov (add twitter)
Ask The Authors

Ask the authors of this paper a question or leave a comment.

Read it. Rate it.
#1. Which part of the paper did you read?

#2. The paper contains new data or analyses that is openly accessible?
#3. The conclusion is supported by the data and analyses?
#4. The conclusion is of scientific interest?
#5. The result is likely to lead to future research?

Github
User:
None (add)
Repo:
None (add)
Stargazers:
0
Forks:
0
Open Issues:
0
Network:
0
Subscribers:
0
Language:
None
Youtube
Link:
None (add)
Views:
0
Likes:
0
Dislikes:
0
Favorites:
0
Comments:
0
Other
Sample Sizes (N=):
Inserted:
Words Total:
Words Unique:
Source:
Abstract:
None
11/06/19 06:04PM
7,820
2,580
Tweets
__isaac_s__: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
soooooo_soooooo: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ApsisInc: Facebook’s latest giant language #AI hits computing wall at 500 Nvidia GPUs - https://t.co/QkMXPyXZNg
esmittramirez: Unsupervised Cross-lingual Representation Learning at Scale by Facebook's AI team. XLM-R ftw https://t.co/61m3AhdRGW
jholston: Still gonna be a while before we get that simultaneous translator amulet ... ... "the XLM-R runs into “resource constraints despite using five hundred of Nvidia’s most powerful GPUs,” which the authors refer to as the “curse of multilinguality.” ... https://t.co/PD9TIzMNLn #AI
ibrahim7rasim: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
nljubesic: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
krsna_deepak: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
qoboty: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
Cncylgms: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
linekin: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
samzakalik: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
alexrives: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
DOinterview: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
tianchez_ai: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ishant30oct: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
agvsmontero: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
KlimZaporojets: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
A_K_Nain: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ampetrossian: XLM-RoBERTa: Amazing results on XLU and GLUE benchmarks from Facebook AI: large transformer network trained on 2.5TB of text from 100 languages. - ArXiv paper: https://t.co/cWfdS2agzr - Gitub repo:... https://t.co/mTkMctHJYQ
delphinel: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
hierocryptiks: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
chainsawriot: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
avillanovamoral: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
syoyo: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
kporterrobinson: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
akpanydre: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
bhagirathl: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
pbcquoc: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
xin_josh: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
racingnull: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
selinplus: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
benbenfiol: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
margusb: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
meapistol: Not a week goes by without dramatic progress in NLP. One net trained on 100 languages beats monolingual models in many cases. Transformers rock: https://t.co/jtAM091WEW
kl4sp: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
MishakinSergey: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
TravisRPetersen: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
DieEmme: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
HARSH_306: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
juandaserni: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
FlemmingMa: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
boringsegfault: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
dave_co_dev: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
stanfordnlp: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
BFelbo: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
eisenzopf: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
tadejtadej: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
singhkrrohit: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
rossgritz: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
unsorsodicorda: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
mkaipetersen: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
morinosuke__p: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
debora_nozza: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
luyiwang: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
JamesReilly273: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
sumitsethy: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
jeandut14000: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
morgangiraud: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
CBobed: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
woc_sajit: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
A_Couloumy: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
jd_mashiro: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
PerthMLGroup: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
evolvingstuff: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
AssistedEvolve: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
altamborrino: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
rose_miura: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
DocXavi: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
migballesteros: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ryo_masumura: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
Sam09lol: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
wilderrodrigues: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
xatkit: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
eisenjulian: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
PierreStock: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
SorishaPragyan: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ChristianKlose3: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
weballergy: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
mahesh21aug: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
giroxarmend: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
NoppadonKoo: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ParisMLgroup: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
KouroshMeshgi: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
sanjaykamath: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
yoquankara: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
BramVanroy: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ballforest: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
__MLT__: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
DeepMatrixCloud: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
talkdatatomee: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
smart_grid_fr: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
sbmaruf: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
mgrankin: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
HrSaghir: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
panteon_college: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
EGenus7: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
deepak_s_rawat: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
LukeZettlemoyer: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
tmramalho: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
dannyehb: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
hudsonmendes: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
pijili: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
Charles9n: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
flowing: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
guzmanhe: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ogrisel: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
carlos_g_ospino: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
akkikiki: "for 1.5 Million updates on five hundred 32GB Nvidia V100 GPUs with a batch size of 8192" https://t.co/c1skn1UAZm
raiyani: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
liucaoming: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
pragmaticml: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
DannyAmadoL: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
Tim_Dettmers: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
bnjasim: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
_jongwook_kim: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
Didar_Olmez: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
zdepablo: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
gsarti_: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
ThawabHAlsubaie: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
v_trokhymenko: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
arxivml: "Unsupervised Cross-lingual Representation Learning at Scale", Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vi… https://t.co/3BMj0Zaza1
jose_e_pons: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
phtully: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
yuvalmarton: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
roeeaharoni: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
tiagomluis: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
GillesMoyse: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
andyhickl: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
numPerfecto28: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
michalwols: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
WonderMicky: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
anne_lauscher: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
tuvuumass: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
stormtroper1721: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
kalpeshk2011: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
louismrt: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
loretoparisi: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
julien_c: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
Kelechukwu_: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
snisarg: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
shrimai_: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
VishravC: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
daniellevy__: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
EzraWu: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
NamanGoyal21: RT @alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a T…
alex_conneau: Our new paper: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/N5nTKhUBnE We release XLM-R, a Transformer MLM trained in 100 langs on 2.5 TB of text data. Double digit gains on XLU benchmarks + strong per-language performance (~XLNet on GLUE). [1/6] https://t.co/0RX1ljGuri
ukhndlwl: RT @kakemeister: You can find the paper here: https://t.co/HVPjHdtxuS
kakemeister: @alex_conneau @facebookai @NamanGoyal21 @VishravC @gwenzek @guzmanhe @EXGRV @myleott @LukeZettlemoyer @vesko_st You can find the paper here: https://t.co/HVPjHdtxuS
kakemeister: You can find the paper here: https://t.co/HVPjHdtxuS
arxiv_cscl: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/PjcMfX07iA
_stefan_munich: "Unsupervised Cross-lingual Representation Learning at Scale" is out now: https://t.co/8rlc193oO4 https://t.co/rBw4Myg4qU
MishakinSergey: RT @roadrunning01: Unsupervised Cross-lingual Representation Learning at Scale pdf: https://t.co/RumPI4ORh0 abs: https://t.co/WwlU5mmvXz ht…
MassBassLol: RT @roadrunning01: Unsupervised Cross-lingual Representation Learning at Scale pdf: https://t.co/RumPI4ORh0 abs: https://t.co/WwlU5mmvXz ht…
Pol09122455: RT @roadrunning01: Unsupervised Cross-lingual Representation Learning at Scale pdf: https://t.co/RumPI4ORh0 abs: https://t.co/WwlU5mmvXz ht…
EricSchles: RT @roadrunning01: Unsupervised Cross-lingual Representation Learning at Scale pdf: https://t.co/RumPI4ORh0 abs: https://t.co/WwlU5mmvXz ht…
arxiv_cscl: Unsupervised Cross-lingual Representation Learning at Scale https://t.co/PjcMfX07iA
roadrunning01: Unsupervised Cross-lingual Representation Learning at Scale pdf: https://t.co/RumPI4ORh0 abs: https://t.co/WwlU5mmvXz https://t.co/l5XiXJrBsZ
Images
Related