Bridging Text and Video: A Universal Multimodal Transformer for Video-Audio Scene-Aware Dialog
This addresses the problem of generating informative and fluent dialogue responses for video-based conversational systems, though it appears incremental as it extends existing pre-trained models to a multimodal task.
The paper tackled the Audio-Visual Scene-Aware Dialog (AVSD) task by proposing a universal multimodal transformer with multi-task learning to generate responses about videos, achieving the best performance in objective and subjective evaluations in the DSTC8 challenge.
Audio-Visual Scene-Aware Dialog (AVSD) is a task to generate responses when chatting about a given video, which is organized as a track of the 8th Dialog System Technology Challenge (DSTC8). To solve the task, we propose a universal multimodal transformer and introduce the multi-task learning method to learn joint representations among different modalities as well as generate informative and fluent responses. Our method extends the natural language generation pre-trained model to multimodal dialogue generation task. Our system achieves the best performance in both objective and subjective evaluations in the challenge.