Dataset Card for GEM/indonlg
Link to Main Data Card
You can find the main data card on the GEM Website.
Dataset Summary
IndoNLG is a collection of various Indonesian, Javanese, and Sundanese NLG tasks including summarization, question answering, chit-chat, and three different pairs of machine translation (MT) tasks.
You can load the dataset via:
import datasets
data = datasets.load_dataset('GEM/indonlg')
The data loader can be found here.
website
paper
authors
Samuel Cahyawijaya, Genta Indra Winata, Bryan Wilie, Karissa Vincentio, Xiaohong Li, Adhiguna Kuncoro, Sebastian Ruder, Zhi Yuan Lim, Syafri Bahar, Masayu Leylia Khodra, Ayu Purwarianti, Pascale Fung
Dataset Overview
Where to find the Data and its Documentation
Webpage
Download
Paper
BibTex
@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. ',}
Contact Name
Genta Indra Winata
Contact Email
Has a Leaderboard?
no
Languages and Intended Use
Multilingual?
yes
Covered Languages
Indonesian, Javanese, Sundanese
License
mit: MIT License
Intended Use
IndoNLG is a collection of Natural Language Generation (NLG) resources for Bahasa Indonesia with 10 downstream tasks.
Primary Task
Summarization
Communicative Goal
Generate a response according to the context and text.
Credit
Curation Organization Type(s)
academic, industry
Curation Organization(s)
The Hong Kong University of Science and Technology, Gojek, Institut Teknologi Bandung, Universitas Multimedia Nusantara, DeepMind, Prosa.ai
Dataset Creators
Samuel Cahyawijaya, Genta Indra Winata, Bryan Wilie, Karissa Vincentio, Xiaohong Li, Adhiguna Kuncoro, Sebastian Ruder, Zhi Yuan Lim, Syafri Bahar, Masayu Leylia Khodra, Ayu Purwarianti, Pascale Fung
Funding
The Hong Kong University of Science and Technology, Gojek, Institut Teknologi Bandung, Universitas Multimedia Nusantara, DeepMind, Prosa.ai
Who added the Dataset to GEM?
Genta Indra Winata (The Hong Kong University of Science and Technology)
Dataset Structure
Dataset in GEM
Rationale for Inclusion in GEM
Similar Datasets
yes
Unique Language Coverage
no
GEM-Specific Curation
Modificatied for GEM?
yes
GEM Modifications
other
Additional Splits?
no
Getting Started with the Task
Previous Results
Previous Results
Measured Model Abilities
Dialog understanding, summarization, translation
Metrics
BLEU
Proposed Evaluation
BLEU evaluates the generation quality.
Previous results available?
yes
Other Evaluation Approaches
BLEU
Dataset Curation
Original Curation
Sourced from Different Sources
no
Language Data
How was Language Data Obtained?
Crowdsourced
Where was it crowdsourced?
Participatory experiment
Data Validation
validated by data curator
Was Data Filtered?
not filtered
Structured Annotations
Additional Annotations?
none
Annotation Service?
no
Consent
Any Consent Policy?
yes
Consent Policy Details
Annotators agree using the dataset for research purpose.
Other Consented Downstream Use
Any
Private Identifying Information (PII)
Contains PII?
unlikely
Categories of PII
``
Maintenance
Any Maintenance Plan?
no
Broader Social Context
Previous Work on the Social Impact of the Dataset
Usage of Models based on the Data
no
Impact on Under-Served Communities
Addresses needs of underserved Communities?
yes
Discussion of Biases
Any Documented Social Biases?
no
Considerations for Using the Data
PII Risks and Liability
Potential PII Risk
No
Licenses
Copyright Restrictions on the Dataset
open license
Copyright Restrictions on the Language Data
open license
Known Technical Limitations
- Downloads last month
- 29