You need to share your contact information to access this dataset.

This repository is publicly accessible, but you have to register to access its content — don't worry, it's just one click!

By clicking on “Access repository” below, you accept that your contact information (email address and username) can be shared with the repository authors. This will let the authors get in touch for instance if some parts of the repository's contents need to be taken down for licensing reasons.

You will immediately be granted access to the contents of the dataset.

Access repository

Dataset Card for "indonlg"

Dataset Summary

IndoNLG is a collection of Natural Language Generation (NLG) resources for Indonesian, Javanese, and Sundanese with 6 kind of downstream tasks.

Supported Tasks and Leaderboards

More Information Needed

Languages

Indonesian, Javanese, Sundanese

Dataset Structure

We show detailed information for 10 datasets.

Data Instances

indosum

Size of downloaded dataset files: 50.5 MB
Size of the generated dataset: 50.5 MB
Total amount of disk used: 50.5 MB

liputan6_canonical

(Liputan6 (Canonical))

Size of downloaded dataset files: 363.6 MB
Size of the generated dataset: 363.6 MB
Total amount of disk used: 363.6 MB

liputan6_xtreme

(Liputan6 (Xtreme))

Size of downloaded dataset files: 311.5 MB
Size of the generated dataset: 311.5 MB
Total amount of disk used: 311.5 MB

MT_ENGKJV_INZNTV

(Bible En↔Id)

Size of downloaded dataset files: 10.2 MB
Size of the generated dataset: 10.2 MB
Total amount of disk used: 10.2 MB

MT_IMD_NEWS

(News En↔Id)

Size of downloaded dataset files: 13.6 MB
Size of the generated dataset: 13.6 MB
Total amount of disk used: 13.6 MB

MT_JAVNRF_INZNTV

(Bible Jv↔Id)

Size of downloaded dataset files: 3.3 MB
Size of the generated dataset: 3.3 MB
Total amount of disk used: 3.3 MB

MT_SUNIBS_INZNTV

(Bible Su↔Id)

Size of downloaded dataset files: 2.6 MB
Size of the generated dataset: 2.6 MB
Total amount of disk used: 2.6 MB

MT_TED_MULTI

(TED En↔Id)

Size of downloaded dataset files: 23.4 MB
Size of the generated dataset: 23.4 MB
Total amount of disk used: 23.4 MB

question_answering

Size of downloaded dataset files: 4.7 MB
Size of the generated dataset: 4.7 MB
Total amount of disk used: 4.7 MB

xpersona

Size of downloaded dataset files: 19.8 MB
Size of the generated dataset: 19.8 MB
Total amount of disk used: 19.8 MB

Data Fields

The data fields are the same among all splits

gem_id: a string feature.
context: a string feature.
input: a string feature.
target: a string feature.
references: a list of string feature.

Data Splits

task	train	validation	test
indosum	14083	1880	2810
liputan6_canonical	193,883	10,972	10,972
liputan6_xtreme	193,883	4,948	3,862
MT_ENGKJV_INZNTV	23,308	3,109	4,661
MT_IMD_NEWS	38,469	1,953	1,954
MT_JAVNRF_INZNTV	14083	1880	2810
MT_SUNIBS_INZNTV	14083	1880	2810
MT_TED_MULTI	14083	1880	2810
question_answering (TyDiQA)	4,847	565	565
xpersona	16,878	484	484

Dataset Creation

Curation Rationale

More Information Needed

Source Data

Initial Data Collection and Normalization

More Information Needed

Who are the source language producers?

More Information Needed

Annotations

Considerations for Using the Data

Social Impact of Dataset

More Information Needed

Discussion of Biases

More Information Needed

Other Known Limitations

More Information Needed

Additional Information

Dataset Curators

More Information Needed

Licensing Information

More Information Needed

Citation Information

@inproceedings{cahyawijaya-etal-2021-indonlg,
    title = "{I}ndo{NLG}: Benchmark and Resources for Evaluating {I}ndonesian Natural Language Generation",
    author = "Cahyawijaya, Samuel  and
      Winata, Genta Indra  and
      Wilie, Bryan  and
      Vincentio, Karissa  and
      Li, Xiaohong  and
      Kuncoro, Adhiguna  and
      Ruder, Sebastian  and
      Lim, Zhi Yuan  and
      Bahar, Syafri  and
      Khodra, Masayu  and
      Purwarianti, Ayu  and
      Fung, Pascale",
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.emnlp-main.699",
    pages = "8875--8898",
    abstract = "Natural language generation (NLG) benchmarks provide an important avenue to measure progress and develop better NLG systems. Unfortunately, the lack of publicly available NLG benchmarks for low-resource languages poses a challenging barrier for building NLG systems that work well for languages with limited amounts of data. Here we introduce IndoNLG, the first benchmark to measure natural language generation (NLG) progress in three low-resource{---}yet widely spoken{---}languages of Indonesia: Indonesian, Javanese, and Sundanese. Altogether, these languages are spoken by more than 100 million native speakers, and hence constitute an important use case of NLG systems today. Concretely, IndoNLG covers six tasks: summarization, question answering, chit-chat, and three different pairs of machine translation (MT) tasks. We collate a clean pretraining corpus of Indonesian, Sundanese, and Javanese datasets, Indo4B-Plus, which is used to pretrain our models: IndoBART and IndoGPT. We show that IndoBART and IndoGPT achieve competitive performance on all tasks{---}despite using only one-fifth the parameters of a larger multilingual model, mBART-large (Liu et al., 2020). This finding emphasizes the importance of pretraining on closely related, localized languages to achieve more efficient learning and faster inference at very low-resource languages like Javanese and Sundanese.",
}