Reasoning Visual Dialogs with Structural and Partial Observations
This addresses the challenge of reasoning in visual dialogs with complex structures, which is incremental as it builds on existing graphical model and GNN approaches.
The paper tackles the problem of Visual Dialog by modeling it as inference in a graphical model with partially observed nodes and unknown structures, proposing an EM algorithm and a differentiable GNN solution. Results show that the model outperforms comparative methods on VisDial and VisDial-Q datasets and can infer underlying dialog structures for better reasoning.
We propose a novel model to address the task of Visual Dialog which exhibits complex dialog structures. To obtain a reasonable answer based on the current question and the dialog history, the underlying semantic dependencies between dialog entities are essential. In this paper, we explicitly formalize this task as inference in a graphical model with partially observed nodes and unknown graph structures (relations in dialog). The given dialog entities are viewed as the observed nodes. The answer to a given question is represented by a node with missing value. We first introduce an Expectation Maximization algorithm to infer both the underlying dialog structures and the missing node values (desired answers). Based on this, we proceed to propose a differentiable graph neural network (GNN) solution that approximates this process. Experiment results on the VisDial and VisDial-Q datasets show that our model outperforms comparative methods. It is also observed that our method can infer the underlying dialog structure for better dialog reasoning.