Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization
This addresses a key limitation in conversational AI for applications like chatbots, though it appears incremental as it builds on existing adversarial and information-theoretic methods.
The paper tackled the problem of neural conversational models producing uninformative and non-diverse responses by introducing Adversarial Information Maximization (AIM), which improved both aspects significantly in evaluations.
Responses generated by neural conversational models tend to lack informativeness and diversity. We present Adversarial Information Maximization (AIM), an adversarial learning strategy that addresses these two related but distinct problems. To foster response diversity, we leverage adversarial training that allows distributional matching of synthetic and real responses. To improve informativeness, our framework explicitly optimizes a variational lower bound on pairwise mutual information between query and response. Empirical results from automatic and human evaluations demonstrate that our methods significantly boost informativeness and diversity.