facebookresearch / flores

Facebook Low Resource (FLoRes) MT Benchmark

facebookresearch/flores Issues

contamination flores200-dev and flores200-devtest
Updated 8 months ago
Adding new languages to FLORES dataset
Updated 9 months ago
English Devtest Line 439: An extra "I" after sentence.
Updated 9 months ago
The translations in Moroccan Arabic (ary_Arab) are just Modern Standard Arabic.
Updated 10 months ago
Standard Moroccan Tamazight mislabeled.
Updated 10 months ago
Wrong Text on Spanish Devset Line 536
Updated a year ago
The Cantonese (Yue Chinese, `yue_Hant`) data in FLORES-200 is not Cantonese at all
Updated a year ago1
Tinyurl.com download for Flores200 gives a certificate error
Closed a year ago
About the function creep and 9 improvements to the file est_Latn_twl.txt
Closed a year ago1
Dataset Problem.
Closed 2 years ago3
None
Closed 2 years ago
Request for the correction of Santali script name
Closed 2 years ago3
Central Kurdish Problems
Updated 2 years ago4
Contribution: Kabyle Language
Closed 2 years ago3
Evaluation Script for all language pairs
Closed 2 years ago
Could not reproduce FloresV1 BLEU scores
Updated 2 years ago
Topics
Updated 2 years ago
Non-matching quotation marks in some dev/devtest sets
Updated 2 years ago
Scope for addition of New Language Bodo
Closed 2 years ago2
Data Download is outdated
Updated 3 years ago
Extra zero-width characters in the dataset
Updated 3 years ago
Mismatch of the size between pretrained model and finetuned model
Closed 3 years ago1
RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'
Updated 3 years ago
FLORES-101 benchmark and Alternative Spelling rules in some languages, using FLORES-101 for benchmarking embeddings models
Updated 3 years ago2
Submission to DynaBench does not show results
Closed 3 years ago1
Question about the x->traditional Chinese
Updated 3 years ago
Evaluation of languages of the same family
Closed 3 years ago1
Problems in Catalan files (encoding / conversions)
Closed 3 years ago4
Pashto Language does not have test set
Closed 3 years ago1
Scripts to Scrape Data?
Closed 3 years ago5
200k Sinhala parallel sentences are filtered
Closed 3 years ago2
Not able to reproduce the semi-supervised results
Closed 3 years ago1
questions about unsupervised results in the paper
Closed 3 years ago
Broken download link
Closed 3 years ago2
detokenize output
Closed 3 years ago1
How to replicate supervised NE-EN baseline?
Closed 4 years ago2
ERROR in download-data.sh
Closed 5 years ago5
Scripts for Back-translation?
Closed 5 years ago3
Can't Decompress ”commoncrawl.deduped.en.xz“.
Closed 5 years ago1
Is monolingual data used in the paper available for downloading?
Closed 5 years ago4
Reproducing Fully Supervised Baseline
Closed 5 years ago7
bad gateway for data downloading
Closed 5 years ago2
how to get the monolingual corpus
Closed 5 years ago
Error encountered when running prepare-neen.sh
Closed 5 years ago2