Efficient-Empathy: Towards Efficient and Effective Selection of Empathy Data

TL;DR

Efficient-Empathy introduces a data selection algorithm that improves empathetic response quality in large language models by selecting high-quality sensibility and rationality data, leading to state-of-the-art performance with less data.

Contribution

The paper presents a novel sensibility and rationality score-based data selection method that enhances empathetic dialogue models' efficiency and effectiveness.

Findings

Using only 59% of data, achieves state-of-the-art performance.

Data selection hyperparameters maintain high performance.

Combining sensibility and rationality data with MoE further improves results.

Abstract

In recent years, with the rapid advancements in large language models (LLMs), achieving excellent empathetic response capability has become a crucial prerequisite. Consequently, managing and understanding large-scale video datasets has gained increasing importance. However, empathetic data are typically trained without any quality selection, leading to inefficient data usage and wasted computational resources. Additionally, using raw data can result in low performance in empathetic dialogues. In this work, we present Efficient-Empathy, a sensibility and rationality score-based data selection algorithm that automatically selects sensibility and rationality data while discarding low-quality data. With only the sensibility data (59% of the full dataset), our trained sensibility model efficiently achieves state-of-the-art (SoTA) performance. Furthermore, with multiple data selection hyperparameters, the sensibility model demonstrates SoTA performance, showcasing the robustness of our method. By integrating sensibility and rationality data with a MoE structure, we achieve even higher performance, demonstrating the effectiveness of our Efficient-Empathy algorithm.