Abstract
The diversity and complexity of Indic languages present unique challenges fornatural language processing (NLP) tasks, particularly in the domain of questionanswering (QA).To address these challenges, this paper explores the applicationof State Space Models (SSMs),to build efficient and contextually aware QAsystems tailored for Indic languages. SSMs are particularly suited for thistask due to their ability to model long-term and short-term dependencies insequential data, making them well-equipped to handle the rich morphology,complex syntax, and contextual intricacies characteristic of Indian languages.We evaluated multiple SSM architectures across diverse datasets representingvarious Indic languages and conducted a comparative analysis of theirperformance. Our results demonstrate that these models effectively capturelinguistic subtleties, leading to significant improvements in questioninterpretation, context alignment, and answer generation. This work representsthe first application of SSMs to question answering tasks in Indic languages,establishing a foundational benchmark for future research in this domain. Wepropose enhancements to existing SSM frameworks, optimizing their applicabilityto low-resource settings and multilingual scenarios prevalent in Indiclanguages.