Learning from Relevant Subgoals in Successful Dialogs using Iterative Training for Task-oriented Dialog Systems

2024/11/25 by Magdalena Kaiser, Kaiser, Magdalena, Patrick Erñst +3 · 1 citation
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multi-Agent Systems and Negotiation #Speech and dialogue systems #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2411.16305

openalex publication_date 2024/11/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Task-oriented Dialog (ToD) systems have to solve multiple subgoals to accomplish user goals, whereas feedback is often obtained only at the end of the dialog. In this work, we propose SUIT (SUbgoal-aware ITerative Training), an iterative training approach for improving ToD systems. We sample dialogs from the model we aim to improve and determine subgoals that contribute to dialog success using distant supervision to obtain high quality training samples. We show how this data improves supervised fine-tuning or, alternatively, preference learning results. SUIT is able to iteratively generate more data instead of relying on fixed static sets. SUIT reaches new state-of-the-art performance on a popular ToD benchmark.

Cited by

• Empowering LLMs in Task-Oriented Dialogues: A Domain-Independent Multi-Agent Framework and Fine-Tuning Strategy

Related

• Self-training Improves Pre-training for Few-shot Learning in Task-oriented Dialog Systems
• GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection
• Iterative Policy Learning in End-to-End Trainable Task-Oriented Neural Dialog Models
• End-to-end LSTM-based dialog control optimized with supervised and reinforcement learning
• Semi-Supervised Knowledge-Grounded Pre-training for Task-Oriented Dialog Systems