Improving Dialog Systems for Negotiation with Personality Modeling
This work addresses negotiation dialog systems for AI agents, presenting an incremental improvement through personality modeling.
The paper tackled the problem of improving dialog systems for negotiation by modeling opponent personality types to adapt strategies, achieving a 20% higher dialog agreement rate compared to baselines on the CraigslistBargain dataset.
In this paper, we explore the ability to model and infer personality types of opponents, predict their responses, and use this information to adapt a dialog agent's high-level strategy in negotiation tasks. Inspired by the idea of incorporating a theory of mind (ToM) into machines, we introduce a probabilistic formulation to encapsulate the opponent's personality type during both learning and inference. We test our approach on the CraigslistBargain dataset and show that our method using ToM inference achieves a 20% higher dialog agreement rate compared to baselines on a mixed population of opponents. We also find that our model displays diverse negotiation behavior with different types of opponents.