FluGCF: A Fluent Dialogue Generation Model With Coherent Concept Entity Flow

计算机科学 流利 对话 词汇 自然语言处理 人工智能 判决 语言学 哲学
作者
Yu Zhao,Bo Cheng,Yunte Huang,Zhiguo Wan
出处
期刊:IEEE/ACM transactions on audio, speech, and language processing [Institute of Electrical and Electronics Engineers]
卷期号:32: 853-867
标识
DOI:10.1109/taslp.2023.3340610
摘要

The integration of external knowledge graphs into dialogue systems effectively mitigates the generation of generic and uninteresting responses. This approach, particularly the explicit modeling of conversation flows from related concept entities, facilitates the generation of semantically rich and informative responses. However, recent models guided by concept entity flows present two primary limitations: (1) a limited semantic understanding of the post message, which complicates the selection of highly relevant 1-hop concept entities, and (2) an inability to extract dynamic and diverse semantic relations between the post message and 2-hop concept entities. To address these issues, we introduce FluGCF, a novel model that fluently generates dialogues with coherent guidance from concept entity flows. FluGCF employs a ternary fusion to explicitly model multi-hop concept entity flows using a post-aware knowledge encoding mechanism. This mechanism learns semantic concept entity features from both word and sentence-level text features. Additionally, we design a corresponding ternary decoding mechanism that dynamically selects concept entities or words from the vocabulary to enhance fluency and diversity in dialogue generation. FluGCF, implemented in PyTorch, was extensively evaluated on a large-scale dataset, revealing that it surpasses baseline models, including the state-of-the-art knowledge-aware model ConceptFlow, by nearly 15% in terms of fluency. Furthermore, it demonstrated notable enhancements in coherence, diversity and informativeness.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
猪猪hero应助123456采纳,获得10
1秒前
1秒前
Ly发布了新的文献求助20
1秒前
cwm完成签到,获得积分10
1秒前
微笑的依凝完成签到,获得积分10
2秒前
book完成签到,获得积分10
2秒前
整齐思天完成签到,获得积分10
2秒前
3秒前
3秒前
shinn发布了新的文献求助10
3秒前
Lucas应助Roseaiwade采纳,获得10
4秒前
好好毕业完成签到,获得积分20
5秒前
风趣的胜应助zj采纳,获得10
6秒前
英姑应助无梦为安采纳,获得10
6秒前
7秒前
科研泥猴桃完成签到,获得积分10
8秒前
梁三柏应助哭泣的凡英采纳,获得10
8秒前
8秒前
苗条的小肥羊完成签到,获得积分10
9秒前
10秒前
研友_LkYoRZ完成签到,获得积分10
12秒前
12秒前
ding应助小新没蜡笔采纳,获得10
13秒前
13秒前
14秒前
14秒前
14秒前
量子星尘发布了新的文献求助10
15秒前
15秒前
15秒前
Roseaiwade发布了新的文献求助10
16秒前
羡鱼完成签到,获得积分10
17秒前
A2QD发布了新的文献求助10
17秒前
18秒前
19秒前
71发布了新的文献求助20
19秒前
19秒前
TQ完成签到,获得积分10
19秒前
彭于晏应助可耐的青雪采纳,获得10
20秒前
思源应助让地球种满香菜采纳,获得10
21秒前
高分求助中
Picture Books with Same-sex Parented Families: Unintentional Censorship 700
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
Effective Learning and Mental Wellbeing 300
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3975165
求助须知:如何正确求助?哪些是违规求助? 3519595
关于积分的说明 11198781
捐赠科研通 3255912
什么是DOI,文献DOI怎么找? 1798001
邀请新用户注册赠送积分活动 877343
科研通“疑难数据库(出版商)”最低求助积分说明 806298