MILU: A Multi-task Indic Language Understanding Benchmark

TL;DR

MILU is a comprehensive benchmark designed to evaluate large language models on 11 Indic languages across diverse domains, highlighting current models' struggles with culturally specific knowledge and low-resource languages.

Contribution

Introduces MILU, the first extensive multi-task benchmark for Indic languages, covering 8 domains and 41 subjects, to assess LLMs' cultural and linguistic understanding.

Findings

Open multilingual models outperform language-specific fine-tuned models.

Current LLMs struggle with culturally relevant areas like Arts and Humanities.

Models perform better in high-resource languages than in low-resource ones.

Abstract

Evaluating Large Language Models (LLMs) in low-resource and linguistically diverse languages remains a significant challenge in NLP, particularly for languages using non-Latin scripts like those spoken in India. Existing benchmarks predominantly focus on English, leaving substantial gaps in assessing LLM capabilities in these languages. We introduce MILU, a Multi task Indic Language Understanding Benchmark, a comprehensive evaluation benchmark designed to address this gap. MILU spans 8 domains and 41 subjects across 11 Indic languages, reflecting both general and culturally specific knowledge. With an India-centric design, incorporates material from regional and state-level examinations, covering topics such as local history, arts, festivals, and laws, alongside standard subjects like science and mathematics. We evaluate over 42 LLMs, and find that current LLMs struggle with MILU, with GPT-4o achieving the highest average accuracy at 74 percent. Open multilingual models outperform language-specific fine-tuned models, which perform only slightly better than random baselines. Models also perform better in high resource languages as compared to low resource ones. Domain-wise analysis indicates that models perform poorly in culturally relevant areas like Arts and Humanities, Law and Governance compared to general fields like STEM. To the best of our knowledge, MILU is the first of its kind benchmark focused on Indic languages, serving as a crucial step towards comprehensive cultural evaluation. All code, benchmarks, and artifacts are publicly available to foster open research.