IP-Dialog: Evaluating Implicit Personalization in Dialogue Systems with Synthetic Data

TL;DR

This paper introduces IP-Dialog, a synthetic dataset and evaluation framework for implicit personalization in dialogue systems, enabling better assessment of models' ability to infer user backgrounds without privacy concerns.

Contribution

It presents a novel synthetic data generation method, a comprehensive benchmark with multiple tasks and attributes, and a systematic evaluation framework for implicit personalization.

Findings

The dataset covers 10 tasks and 12 user attributes.

The evaluation framework includes four metrics for attribute awareness and reasoning.

Experiments demonstrate the dataset's reliability and insights into model capabilities.

Abstract

In modern dialogue systems, the ability to implicitly infer user backgrounds from conversations and leverage this information for personalized assistance is crucial. However, the scarcity of high-quality data remains a fundamental challenge to evaluating and improving this capability. Traditional dataset construction methods are labor-intensive, resource-demanding, and raise privacy concerns. To address these issues, we propose a novel approach for automatic synthetic data generation and introduce the Implicit Personalized Dialogue (IP-Dialog) benchmark along with a training dataset, covering 10 tasks and 12 user attribute types. Additionally, we develop a systematic evaluation framework with four metrics to assess both attribute awareness and reasoning capabilities. We further propose five causal graphs to elucidate models' reasoning pathways during implicit personalization. Extensive experiments yield insightful observations and prove the reliability of our dataset.