Community Regularization of Visually-Grounded Dialog
This addresses the challenge of making AI agents communicate more naturally and effectively in visually grounded tasks, though it is incremental in applying social-inspired regularization to existing dialog models.
The paper tackles the problem of learning goal-oriented cooperative dialog between autonomous agents about visual scenes, proposing a multi-agent community-based framework that enforces regularization to improve dialog relevance and coherence without compromising task performance.
The task of conducting visually grounded dialog involves learning goal-oriented cooperative dialog between autonomous agents who exchange information about a scene through several rounds of questions and answers in natural language. We posit that requiring artificial agents to adhere to the rules of human language, while also requiring them to maximize information exchange through dialog is an ill-posed problem. We observe that humans do not stray from a common language because they are social creatures who live in communities, and have to communicate with many people everyday, so it is far easier to stick to a common language even at the cost of some efficiency loss. Using this as inspiration, we propose and evaluate a multi-agent community-based dialog framework where each agent interacts with, and learns from, multiple agents, and show that this community-enforced regularization results in more relevant and coherent dialog (as judged by human evaluators) without sacrificing task performance (as judged by quantitative metrics).