NusaAksara: A Multimodal and Multilingual Benchmark for Preserving Indonesian Indigenous Scripts

TL;DR

NusaAksara introduces a comprehensive benchmark for Indonesian indigenous scripts, combining text and image tasks across multiple languages and scripts, revealing significant challenges in current NLP technologies for low-resource and unsupported scripts.

Contribution

The paper presents NusaAksara, a novel, publicly available benchmark covering diverse Indonesian scripts and languages, including low-resource and unsupported scripts like Lampung, with extensive multimodal tasks and expert-constructed data.

Findings

Most NLP models perform near zero on local Indonesian scripts.

The benchmark includes 8 scripts across 7 languages, including low-resource ones.

Current technologies struggle with unsupported scripts like Lampung.

Abstract

Indonesia is rich in languages and scripts. However, most NLP progress has been made using romanized text. In this paper, we present NusaAksara, a novel public benchmark for Indonesian languages that includes their original scripts. Our benchmark covers both text and image modalities and encompasses diverse tasks such as image segmentation, OCR, transliteration, translation, and language identification. Our data is constructed by human experts through rigorous steps. NusaAksara covers 8 scripts across 7 languages, including low-resource languages not commonly seen in NLP benchmarks. Although unsupported by Unicode, the Lampung script is included in this dataset. We benchmark our data across several models, from LLMs and VLMs such as GPT-4o, Llama 3.2, and Aya 23 to task-specific systems such as PP-OCR and LangID, and show that most NLP technologies cannot handle Indonesia's local scripts, with many achieving near-zero performance.