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

2026-01-09

International Conference on Computational Linguistics

AI Summary

This work addresses the limitations of existing long-term dialogue systems, which often rely on simplistic accumulation or operational approaches for memory updating and struggle to resolve conflicting information or accurately track user states. To overcome these challenges, we propose a generative memory update mechanism that, for the first time, integrates emotional context and causal relationships into the dynamic memory integration process, enabling more precise user state representations. To support this research, we introduce the KEEM dataset. Experimental results demonstrate that our approach significantly enhances the system’s understanding of user states, leading to more empathetic and semantically coherent dialogues, and generating more natural and meaningful responses in open-domain conversations.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Conversational AI/Dialog Systems
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search

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.

Problem

• long-term conversation
• memory update
• emotional context
• information conflict
• user state tracking

Innovation

• generation-based memory
• emotion-reflected memory
• long-term conversation
• KEEM dataset
• integrative memory update