BEDA: Belief Estimation as Probabilistic Constraints for Performing Strategic Dialogue Acts
This addresses the challenge of principled belief utilization in strategic dialogue for AI agents, representing an incremental advance over prior methods.
The paper tackled the problem of using belief estimation to improve strategic dialogue acts by formalizing adversarial and alignment acts as probabilistic constraints, resulting in BEDA, which improved success rates by at least 5.0 points on CKBG, 9.3 points on Mutual Friends, and achieved optimal deals on CaSiNo compared to baselines.
Strategic dialogue requires agents to execute distinct dialogue acts, for which belief estimation is essential. While prior work often estimates beliefs accurately, it lacks a principled mechanism to use those beliefs during generation. We bridge this gap by first formalizing two core acts Adversarial and Alignment, and by operationalizing them via probabilistic constraints on what an agent may generate. We instantiate this idea in BEDA, a framework that consists of the world set, the belief estimator for belief estimation, and the conditional generator that selects acts and realizes utterances consistent with the inferred beliefs. Across three settings, Conditional Keeper Burglar (CKBG, adversarial), Mutual Friends (MF, cooperative), and CaSiNo (negotiation), BEDA consistently outperforms strong baselines: on CKBG it improves success rate by at least 5.0 points across backbones and by 20.6 points with GPT-4.1-nano; on Mutual Friends it achieves an average improvement of 9.3 points; and on CaSiNo it achieves the optimal deal relative to all baselines. These results indicate that casting belief estimation as constraints provides a simple, general mechanism for reliable strategic dialogue.