Synergizing In-context Learning with Hints for End-to-end Task-oriented Dialog Systems

TL;DR

This paper introduces SyncTOD, a method that combines large language models with task-specific hints and auxiliary models to improve task-oriented dialog systems, especially when training data is limited.

Contribution

SyncTOD innovatively integrates hints and exemplar selection with LLMs, enhancing low-data performance and response alignment in dialog systems.

Findings

SyncTOD outperforms baseline LLMs and SOTA models in low-data scenarios.

SyncTOD maintains competitive performance with full data.

Auxiliary models effectively provide hints and select exemplars.

Abstract

End-to-end Task-Oriented Dialog (TOD) systems typically require extensive training datasets to perform well. In contrast, large language model (LLM) based TOD systems can excel even with limited data due to their ability to learn tasks through in-context exemplars. However, these models lack alignment with the style of responses in training data and often generate comprehensive responses, making it difficult for users to grasp the information quickly. In response, we propose SyncTOD that synergizes LLMs with task-specific hints to improve alignment in low-data settings. SyncTOD employs small auxiliary models to provide hints and select exemplars for in-context prompts. With ChatGPT, SyncTOD achieves superior performance compared to LLM-based baselines and SoTA models in low-data settings, while retaining competitive performance in full-data settings.