ChitroJera: A Regionally Relevant Visual Question Answering Dataset for Bangla
This addresses the problem of low-resource VQA for Bangla speakers by providing a locally relevant dataset, though it is incremental as it adapts existing methods to a new domain.
The authors tackled the lack of regionally relevant benchmarks for Bangla visual question answering by introducing ChitroJera, a large-scale dataset with over 15k samples, and found that pre-trained dual-encoders and large vision language models achieved the best performance.
Visual Question Answer (VQA) poses the problem of answering a natural language question about a visual context. Bangla, despite being a widely spoken language, is considered low-resource in the realm of VQA due to the lack of proper benchmarks, challenging models known to be performant in other languages. Furthermore, existing Bangla VQA datasets offer little regional relevance and are largely adapted from their foreign counterparts. To address these challenges, we introduce a large-scale Bangla VQA dataset, ChitroJera, totaling over 15k samples from diverse and locally relevant data sources. We assess the performance of text encoders, image encoders, multimodal models, and our novel dual-encoder models. The experiments reveal that the pre-trained dual-encoders outperform other models of their scale. We also evaluate the performance of current large vision language models (LVLMs) using prompt-based techniques, achieving the overall best performance. Given the underdeveloped state of existing datasets, we envision ChitroJera expanding the scope of Vision-Language tasks in Bangla.