THEANINE：在長期對話中重新審視記憶管理與時間軸增強回應生成

THEANINE: Revisiting Memory Management in Long-term Conversations with Timeline-augmented Response Generation

June 16, 2024

作者: Seo Hyun Kim, Kai Tzu-iunn Ong, Taeyoon Kwon, Namyoung Kim, Keummin Ka, SeongHyeon Bae, Yohan Jo, Seung-won Hwang, Dongha Lee, Jinyoung Yeo

cs.AI

摘要

大型語言模型（LLMs）能夠在與用戶進行長時間互動時處理冗長的對話歷史，而無需額外的記憶模塊；然而，它們的回應往往會忽略或不正確地回憶過去的信息。在本文中，我們重新探討了在LLMs時代的記憶增強式回應生成。儘管先前的工作著重於消除過時的記憶，但我們認為這些記憶可以提供上下文線索，幫助對話系統理解過去事件的發展，因此有助於回應生成。我們提出了Theanine，一個框架，它通過記憶時間軸來增強LLMs的回應生成，這些時間軸展示了相關過去事件的發展和因果關係。除了Theanine之外，我們還介紹了TeaFarm，一個以反事實驅動的問答管道，解決了長期對話中G-Eval的限制。我們的方法的補充視頻和TeaFarm評估的TeaBag數據集可在https://theanine-693b0.web.app/找到。

English

Large language models (LLMs) are capable of processing lengthy dialogue histories during prolonged interaction with users without additional memory modules; however, their responses tend to overlook or incorrectly recall information from the past. In this paper, we revisit memory-augmented response generation in the era of LLMs. While prior work focuses on getting rid of outdated memories, we argue that such memories can provide contextual cues that help dialogue systems understand the development of past events and, therefore, benefit response generation. We present Theanine, a framework that augments LLMs' response generation with memory timelines -- series of memories that demonstrate the development and causality of relevant past events. Along with Theanine, we introduce TeaFarm, a counterfactual-driven question-answering pipeline addressing the limitation of G-Eval in long-term conversations. Supplementary videos of our methods and the TeaBag dataset for TeaFarm evaluation are in https://theanine-693b0.web.app/.