Uncertainty as a Planning Signal: Multi-Turn Decision Making for Goal-Oriented Conversation
This work addresses the problem of poor coordination between information acquisition and target commitment in goal-oriented conversational systems for users and developers, representing an incremental improvement over existing methods.
The paper tackles the challenge of multi-turn decision making in goal-oriented conversational systems by formulating it as an uncertainty-aware sequential decision problem, where uncertainty guides planning to balance information acquisition and target commitment. The proposed CUP framework integrates language models with structured planning, resulting in improved success rates and fewer interaction turns across multiple benchmarks.
Goal-oriented conversational systems require making sequential decisions under uncertainty about the user's intent, where the algorithm must balance information acquisition and target commitment over multiple turns. Existing approaches address this challenge from different perspectives: structured methods enable multi-step planning but rely on predefined schemas, while LLM-based approaches support flexible interactions but lack long-horizon decision making, resulting in poor coordination between information acquisition and target commitment. To address this limitation, we formulate goal-oriented conversation as an uncertainty-aware sequential decision problem, where uncertainty serves as a guiding signal for multi-turn decision making. We propose a Conversation Uncertainty-aware Planning framework (CUP) that integrates language models with structured planning: a language model proposes feasible actions, and a planner evaluates their long-term impact on uncertainty reduction. Experiments on multiple conversational benchmarks show that CUP consistently improves success rates while requiring fewer interaction turns. Further analysis demonstrates that uncertainty-aware planning contributes to more efficient information acquisition and earlier confident commitment.