TigerCoder: A Novel Suite of LLMs for Code Generation in Bangla
This addresses the problem of low-resource language support in AI for Bangla-speaking developers, though it is incremental as it adapts existing methods to a new domain.
The paper tackles the underrepresentation of Bangla in code generation by introducing TigerCoder, a suite of LLMs that achieve 11-18% performance gains over existing models.
Despite being the 5th most spoken language, Bangla remains underrepresented in Large Language Models (LLMs), particularly for code generation. This primarily stems from the scarcity of high-quality data to pre-train and/or finetune such models. Hence, we introduce the first dedicated family of Code LLMs for Bangla (1B & 9B). We offer three major contributions: (1) a comprehensive Bangla code instruction datasets for programming domain adaptation; (2) MBPP-Bangla, an evaluation benchmark for Bangla code generation; and (3) the TigerCoder-family of Code LLMs, achieving significant ~11-18% performance gains at Pass@1 over existing multilingual and general-purpose Bangla LLMs. Our findings show that curated, high-quality datasets can overcome limitations of smaller models for low-resource languages. We open-source all resources to advance further Bangla LLM research.