IndoNLG: Benchmark and Resources for Evaluating Indonesian Natural Language Generation
arXiv:2104.08200
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, local languages to achieve more efficient learning and faster inference for very low-resource languages like Javanese and Sundanese.
arXiv commentAccepted in EMNLP 2021, 10 pages
References in corpus (11)
- BERTScore: Evaluating Text Generation with BERT
- CamemBERT: a Tasty French Language Model
- Multilingual Denoising Pre-training for Neural Machine Translation
- MuRIL: Multilingual Representations for Indian Languages
- Multilingual Translation with Extensible Multilingual Pretraining and Finetuning
- ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
- Establishing Baselines for Text Classification in Low-Resource Languages
- Low Resource Neural Machine Translation: A Benchmark for Five African Languages
- Utility is in the Eye of the User: A Critique of NLP Leaderboards
- Cross-Lingual Machine Speech Chain for Javanese, Sundanese, Balinese, and Bataks Speech Recognition and Synthesis
- WangchanBERTa: Pretraining transformer-based Thai Language Models