Generation-Based and Emotion-Reflected Memory Update: Creating the KEEM Dataset for Better Long-Term Conversation

TL;DR

This paper introduces the KEEM dataset to improve long-term conversational memory by dynamically integrating essential facts and emotional context, leading to more empathetic and coherent interactions.

Contribution

The KEEM dataset and generation-based memory update method offer a novel approach to maintaining nuanced, emotionally-aware long-term memory in conversational systems.

Findings

KEEM improves memory accuracy in long-term conversations

Enhanced emotional understanding leads to more empathetic responses

Dynamic memory generation outperforms static accumulation methods

Abstract

In this work, we introduce the Keep Emotional and Essential Memory (KEEM) dataset, a novel generation-based dataset designed to enhance memory updates in long-term conversational systems. Unlike existing approaches that rely on simple accumulation or operation-based methods, which often result in information conflicts and difficulties in accurately tracking a user's current state, KEEM dynamically generates integrative memories. This process not only preserves essential factual information but also incorporates emotional context and causal relationships, enabling a more nuanced understanding of user interactions. By seamlessly updating a system's memory with both emotional and essential data, our approach promotes deeper empathy and enhances the system's ability to respond meaningfully in open-domain conversations.