Planning with Diffusion Models for Target-Oriented Dialogue Systems

• URL: http://arxiv.org/abs/2504.16858v1
• Date: Wed, 23 Apr 2025 16:27:15 GMT
• Title: Planning with Diffusion Models for Target-Oriented Dialogue Systems
• Authors: Hanwen Du, Bo Peng, Xia Ning,
• Abstract summary: We introduce DiffTOD, a dialogue planning framework for non-sequential dialogue planning.<n>DiffTOD formulates dialogue planning as a trajectory generation problem with conditional guidance.<n>We show that DiffTOD can effectively perform non-myopic lookahead exploration and optimize action strategies over a long horizon.
• Score: 5.079888940901933
• License: http://creativecommons.org/licenses/by/4.0/
• 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.

Related papers

• DFlow: Diverse Dialogue Flow Simulation with Large Language Models [16.209331014315463]
  This paper proposes a novel data simulation method designed to enhance the diversity of synthetic dialogues.<n>We generate a task-oriented dialogue dataset comprising 3,886 dialogue flows across 15 different domains.
  arXiv  Detail & Related papers  (2024-10-18T20:35:28Z)
• Dialogue Action Tokens: Steering Language Models in Goal-Directed Dialogue with a Multi-Turn Planner [51.77263363285369]
  We present an approach called Dialogue Action Tokens that adapts language model agents to plan goal-directed dialogues. The core idea is to treat each utterance as an action, thereby converting dialogues into games where existing approaches such as reinforcement learning can be applied.
  arXiv  Detail & Related papers  (2024-06-17T18:01:32Z)
• Target-constrained Bidirectional Planning for Generation of Target-oriented Proactive Dialogue [11.338393954848632]
  We focus on effective dialogue planning for target-oriented dialogue generation. Inspired by decision-making theories in cognitive science, we propose a novel target-constrained bidirectional planning approach. Our algorithms significantly outperform various baseline models.
  arXiv  Detail & Related papers  (2024-03-10T02:14:24Z)
• TOD-Flow: Modeling the Structure of Task-Oriented Dialogues [77.15457469745364]
  We propose a novel approach focusing on inferring the TOD-Flow graph from dialogue data annotated with dialog acts. The inferred TOD-Flow graph can be easily integrated with any dialogue model to improve its prediction performance, transparency, and controllability.
  arXiv  Detail & Related papers  (2023-12-07T20:06:23Z)
• Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents [121.46051697742608]
  We introduce a new dialogue policy planning paradigm to strategize dialogue problems with a tunable language model plug-in named PPDPP. Specifically, we develop a novel training framework to facilitate supervised fine-tuning over available human-annotated data. PPDPP consistently and substantially outperforms existing approaches on three different proactive dialogue applications.
  arXiv  Detail & Related papers  (2023-11-01T03:20:16Z)
• Self-Explanation Prompting Improves Dialogue Understanding in Large Language Models [52.24756457516834]
  We propose a novel "Self-Explanation" prompting strategy to enhance the comprehension abilities of Large Language Models (LLMs) This task-agnostic approach requires the model to analyze each dialogue utterance before task execution, thereby improving performance across various dialogue-centric tasks. Experimental results from six benchmark datasets confirm that our method consistently outperforms other zero-shot prompts and matches or exceeds the efficacy of few-shot prompts.
  arXiv  Detail & Related papers  (2023-09-22T15:41:34Z)
• Dialogue Planning via Brownian Bridge Stochastic Process for Goal-directed Proactive Dialogue [9.99763097964222]
  Goal-directed dialogue systems aim to proactively reach a pre-determined target through multi-turn conversations. Key to achieving this task lies in planning dialogue paths that smoothly and coherently direct conversations towards the target. We propose a coherent dialogue planning approach that uses a process to model the temporal dynamics of dialogue paths.
  arXiv  Detail & Related papers  (2023-05-09T09:28:23Z)
• Enhancing Task Bot Engagement with Synthesized Open-Domain Dialog [89.35658776144638]
  It is essential to build a system that can handle both TOD and ODD and access different knowledge sources. We propose a framework for automatically generating dialogues that combine knowledge-grounded ODDs and TODs in various settings. We introduce a unified model PivotBot that is capable of appropriately adopting TOD and ODD modes and accessing different knowledge sources.
  arXiv  Detail & Related papers  (2022-12-20T05:51:47Z)
• Act-Aware Slot-Value Predicting in Multi-Domain Dialogue State Tracking [5.816391291790977]
  Dialogue state tracking (DST) aims to track human-machine interactions and generate state representations for managing the dialogue. Recent advances in machine reading comprehension predict both categorical and non-categorical types of slots for dialogue state tracking. We formulate and incorporate dialogue acts, and leverage recent advances in machine reading comprehension to predict both categorical and non-categorical types of slots for dialogue state tracking.
  arXiv  Detail & Related papers  (2022-08-04T05:18:30Z)
• Variational Hierarchical Dialog Autoencoder for Dialog State Tracking Data Augmentation [59.174903564894954]
  In this work, we extend this approach to the task of dialog state tracking for goal-oriented dialogs. We propose the Variational Hierarchical Dialog Autoencoder (VHDA) for modeling the complete aspects of goal-oriented dialogs. Experiments on various dialog datasets show that our model improves the downstream dialog trackers' robustness via generative data augmentation.
  arXiv  Detail & Related papers  (2020-01-23T15:34:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.