VLC-BERT: Visual Question Answering With Contextualized Commonsense Knowledge

Ravi Sahithya, Chinchure Aditya, Sigal Leonid, Liao Renjie, Shwartz Vered. Arxiv 2022

[Paper]    
Applications BERT Model Architecture Multimodal Models Pretraining Methods Transformer

There has been a growing interest in solving Visual Question Answering (VQA) tasks that require the model to reason beyond the content present in the image. In this work, we focus on questions that require commonsense reasoning. In contrast to previous methods which inject knowledge from static knowledge bases, we investigate the incorporation of contextualized knowledge using Commonsense Transformer (COMET), an existing knowledge model trained on human-curated knowledge bases. We propose a method to generate, select, and encode external commonsense knowledge alongside visual and textual cues in a new pre-trained Vision-Language-Commonsense transformer model, VLC-BERT. Through our evaluation on the knowledge-intensive OK-VQA and A-OKVQA datasets, we show that VLC-BERT is capable of outperforming existing models that utilize static knowledge bases. Furthermore, through a detailed analysis, we explain which questions benefit, and which don’t, from contextualized commonsense knowledge from COMET.

Similar Work