Generative Visual Dialogue System via Adaptive Reasoning and Weighted Likelihood Estimation
arXiv:1902.09818
Abstract
The key challenge of generative Visual Dialogue (VD) systems is to respond to human queries with informative answers in natural and contiguous conversation flow. Traditional Maximum Likelihood Estimation (MLE)-based methods only learn from positive responses but ignore the negative responses, and consequently tend to yield safe or generic responses. To address this issue, we propose a novel training scheme in conjunction with weighted likelihood estimation (WLE) method. Furthermore, an adaptive multi-modal reasoning module is designed, to accommodate various dialogue scenarios automatically and select relevant information accordingly. The experimental results on the VisDial benchmark demonstrate the superiority of our proposed algorithm over other state-of-the-art approaches, with an improvement of 5.81% on recall@10.
IJCAI 2019
References in corpus (6)
- Sequence to Sequence Learning with Neural Networks
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Adversarial Learning for Neural Dialogue Generation
- A Network-based End-to-End Trainable Task-oriented Dialogue System
- Multi-step Reasoning via Recurrent Dual Attention for Visual Dialog
- FlipDial: A Generative Model for Two-Way Visual Dialogue