Probabilistic Modeling of Intentions in Socially Intelligent LLM Agents
This work addresses the challenge of uncertainty in social interactions for LLM agents, offering an incremental improvement in dialogue performance.
The paper tackles the problem of modeling partner intentions in multi-turn social dialogue for LLM agents, proposing a probabilistic framework that improves Overall scores by 9.0% on SOTOPIA-All and 4.1% on SOTOPIA-Hard compared to a baseline, and slightly surpasses an oracle agent.
We present a probabilistic intent modeling framework for large language model (LLM) agents in multi-turn social dialogue. The framework maintains a belief distribution over a partner's latent intentions, initialized from contextual priors and dynamically updated through likelihood estimation after each utterance. The evolving distribution provides additional contextual grounding for the policy, enabling adaptive dialogue strategies under uncertainty. Preliminary experiments in the SOTOPIA environment show consistent improvements: the proposed framework increases the Overall score by 9.0% on SOTOPIA-All and 4.1% on SOTOPIA-Hard compared with the Qwen2.5-7B baseline, and slightly surpasses an oracle agent that directly observes partner intentions. These early results suggest that probabilistic intent modeling can contribute to the development of socially intelligent LLM agents.