Effect of Visual Extensions on Natural Language Understanding in Vision-and-Language Models
Taichi Iki, Akiko Aizawa · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · 2021
A method for creating a vision-and-language (V&L) model is to extend a language model through structural modifications and V&L pretraining.Such an extension aims to make a V&L model inherit the capability of natural language understanding (NLU) from the original language model.To see how well this is achieved, we propose to evaluate V&L models using an NLU benchmark (GLUE).We compare five V&L models, including singlestream and dual-stream models, trained with the same pre-training.Dual-stream models, with their higher modality independence achieved by approximately doubling the number of parameters, are expected to preserve the NLU capability better.Our main finding is that the dual-stream scores are not much different than the single-stream scores, contrary to expectation.Further analysis shows that pre-training causes the performance drop in NLU tasks with few exceptions.These results suggest that adopting a single-stream structure and devising the pre-training could be an effective method for improving the maintenance of language knowledge in V&L extensions.