Multimodal Dialogue State Tracking By QA Approach with Data Augmentation
This work addresses multimodal dialogue state tracking for researchers in conversational AI, but it is incremental as it builds on existing encoder-decoder frameworks with specific enhancements.
The paper tackles the Audio-Video Scene-Aware Dialogue (AVSD) task by interpreting it as an open-domain question answering problem and proposes a multimodal QA system with a novel data augmentation technique, achieving significant improvements over the baseline on the DSTC7-AVSD dataset.
Recently, a more challenging state tracking task, Audio-Video Scene-Aware Dialogue (AVSD), is catching an increasing amount of attention among researchers. Different from purely text-based dialogue state tracking, the dialogue in AVSD contains a sequence of question-answer pairs about a video and the final answer to the given question requires additional understanding of the video. This paper interprets the AVSD task from an open-domain Question Answering (QA) point of view and proposes a multimodal open-domain QA system to deal with the problem. The proposed QA system uses common encoder-decoder framework with multimodal fusion and attention. Teacher forcing is applied to train a natural language generator. We also propose a new data augmentation approach specifically under QA assumption. Our experiments show that our model and techniques bring significant improvements over the baseline model on the DSTC7-AVSD dataset and demonstrate the potentials of our data augmentation techniques.