Planning with Diffusion Models for Target-Oriented Dialogue Systems

2025-04-23

AI Summary

In task-oriented dialogue (TOD) systems, conventional sequential dialogue planning suffers from error propagation and myopic decision-making. To address this, we propose DiffTOD—the first diffusion-based, non-sequential dialogue planning framework that formulates planning as conditional trajectory generation. We introduce three novel dynamic guidance mechanisms tailored to explicit goals, implicit user intents, and multi-constraint scenarios, enabling long-horizon strategic optimization. Furthermore, we design a structured representation of the dialogue action space to enhance generation controllability and fidelity. Evaluated on three challenging TOD benchmarks, DiffTOD achieves substantial improvements in long-term goal success rate (+12.7% on average), demonstrating strong generalization and cross-scenario adaptability. Our code and data are publicly released.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Multiagent Planning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Target-Oriented Dialogue (TOD) remains a significant challenge in the LLM era, where strategic dialogue planning is crucial for directing conversations toward specific targets. However, existing dialogue planning methods generate dialogue plans in a step-by-step sequential manner, and may suffer from compounding errors and myopic actions. To address these limitations, we introduce a novel dialogue planning framework, DiffTOD, which leverages diffusion models to enable non-sequential dialogue planning. DiffTOD formulates dialogue planning as a trajectory generation problem with conditional guidance, and leverages a diffusion language model to estimate the likelihood of the dialogue trajectory. To optimize the dialogue action strategies, DiffTOD introduces three tailored guidance mechanisms for different target types, offering flexible guidance towards diverse TOD targets at test time. Extensive experiments across three diverse TOD settings show that DiffTOD can effectively perform non-myopic lookahead exploration and optimize action strategies over a long horizon through non-sequential dialogue planning, and demonstrates strong flexibility across complex and diverse dialogue scenarios. Our code and data are accessible through https://anonymous.4open.science/r/DiffTOD.

Problem

• Non-sequential dialogue planning for target-oriented conversations
• Reducing compounding errors in dialogue action strategies
• Flexible guidance for diverse dialogue targets

Innovation

• Leverages diffusion models for non-sequential planning
• Formulates planning as guided trajectory generation
• Uses tailored guidance for diverse target types