Generative Visual Dialogue System via Adaptive Reasoning and Weighted Likelihood Estimation
This work addresses the challenge of generating informative and natural responses in visual dialogue systems, which is an incremental improvement over existing methods.
The paper tackled the problem of generative visual dialogue systems producing generic responses by proposing a weighted likelihood estimation method and adaptive multi-modal reasoning module, achieving a 5.81% improvement in recall@10 on the VisDial benchmark.
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.