Exploiting BERT For Multimodal Target Sentiment Classification Through Input Space Translation

TL;DR

This paper presents a novel two-stream model that translates images into auxiliary text to enhance multimodal sentiment classification on Twitter, achieving state-of-the-art results without modifying language models.

Contribution

Introduces an input space translation approach using object-aware transformers to incorporate visual information into language models for multimodal sentiment analysis.

Findings

Achieves state-of-the-art performance on Twitter datasets.

Effectively distills object-level image information into auxiliary text.

Demonstrates the benefit of input translation over internal multimodal model modifications.

Abstract

Multimodal target/aspect sentiment classification combines multimodal sentiment analysis and aspect/target sentiment classification. The goal of the task is to combine vision and language to understand the sentiment towards a target entity in a sentence. Twitter is an ideal setting for the task because it is inherently multimodal, highly emotional, and affects real world events. However, multimodal tweets are short and accompanied by complex, possibly irrelevant images. We introduce a two-stream model that translates images in input space using an object-aware transformer followed by a single-pass non-autoregressive text generation approach. We then leverage the translation to construct an auxiliary sentence that provides multimodal information to a language model. Our approach increases the amount of text available to the language model and distills the object-level information in complex images. We achieve state-of-the-art performance on two multimodal Twitter datasets without modifying the internals of the language model to accept multimodal data, demonstrating the effectiveness of our translation. In addition, we explain a failure mode of a popular approach for aspect sentiment analysis when applied to tweets. Our code is available at \textcolor{blue}{\url{https://github.com/codezakh/exploiting-BERT-thru-translation}}.