Exploring Effective Information Utilization in Multi-Turn Topic-Driven Conversations

TL;DR

This paper investigates methods to improve information utilization in multi-turn, topic-driven conversations by encoding dialogue history and topics with multi-channel Fusion-in-Decoder, enhancing the handling of long texts in dialogue models.

Contribution

It introduces a multi-channel FiD approach to better fuse dialogue history and topic information, expanding the effective input length for pre-trained language models.

Findings

Multi-channel FiD improves information encoding from long texts.

Combining whole passage and history channels yields competitive results.

Proposed method enhances dialogue generation with extensive context.

Abstract

Conversations are always related to certain topics. However, it is challenging to fuse dialogue history and topic information from various sources at the same time in current dialogue generation models because of the input length limit of pre-trained language models (PLMs). In order to expand the information that PLMs can utilize, we encode topic and dialogue history information using certain prompts with multiple channels of Fusion-in-Decoder (FiD) and explore the influence of three different channel settings. In this paper, our experiments focus on a specific Chinese dataset named NaturalConv, where the conversation revolves around a piece of recent news. We thoroughly compared different dialogue models and different FiD channel settings. Empirical results show that by combining our proposed whole passage channel with additional history channel, our methods can achieve competitive performance on NaturalConv, making it possible to encode various information from excessively long texts.