KnowledgeAwareConversationGenerationReasoningonAugmentedGraphs.pdf资源-CSDN文库

需积分: 50 117 浏览量 2019-08-09 15:11:31 上传评论收藏 390KB PDF 举报

资源推荐

资源详情

资源评论

arXiv:1903.10245v3 [cs.AI] 19 Apr 2019

Knowledge Aware Conversation Generation with Explainable Reasoning

on Augmented Graphs

Zhibin Liu Zheng-Yu Niu Hua Wu Haifeng Wang

Baidu Inc., Beijing, China

{liuzhibin05,niuzhengyu,wu

hua,wanghaifeng}@baidu.com

Abstract

Two types of knowledge, triples from knowl-

edge gr aphs and texts from unstructured doc-

uments, have been studied for knowledge

aware open-domain conversation generation,

in which triple attributes or graph paths can

narrow down vertex candidates for knowledge

selection decision, and texts can provide rich

informa tion for response generation. Fusion of

a knowledge graph and texts might yield mu-

tually reinfor c ing advantages for co nversation

generation, but there is less study on that. To

address this challenge, we propose a knowl-

edge aware cha tting machine with th ree com-

ponen ts, an augmented knowledge graph con-

taining both triples and texts, knowledge selec-

tor, and response generator. For knowledge se-

lection on the graph, we formulate it as a prob-

lem of multi-hop graph reasoning that is more

explainable and ﬂexible in compariso n with

previous works. To fully leverage lon g text

informa tion that differentiates our graph from

others, we improve a state of the art reason-

ing algorithm with machine reading compre-

hension technology. We demonstrate that sup-

ported by such uniﬁed knowledge and explain-

able knowledge selection method, our system

can generate more appropriate and informative

responses than baselines.

1 Introduction

One of the key goals of AI is to build a ma-

chine that can talk with humans when given an

initial topic. To achieve this goal, the machine

should be able to understand language with back-

ground knowledge, recall knowledge from mem-

ory or external resource, reason about these con-

cepts together, and ﬁnally output appropriate and

informative responses. Lots of research efforts

have been devoted to chitchat oriented conversa-

tion generation (

Ritter et al., 2011; Shang et al.,

2015). However, these models tend to produce

generic responses or incoherent responses for a

given topic, since it is quite challenging to learn

semantic interactions merely from dialogue data

(

Ghazvininejad et al., 2018; Zhou et al., 2018a)

without help of background knowledge.

Recently, some previous studies have been con-

ducted to introduce external knowledge in open-

domain chitchat generation (Ghazvininejad et al.,

2018; Liu et al., 2018; Vougiouklis et al., 2016;

Young et al., 2018; Zhou et al., 2018a). These

models usually recall background knowledge

from a source, either unstructured knowledge texts

(

Ghazvininejad et al., 2018; Vougiouklis et al.,

2016) or structured knowledge triples (Liu et al.,

2018; Young et al., 2018; Zhou et al., 2018a),

and then generate more informative responses

conditioned on selected knowledge.

For knowledge triples from a graph, e.g., facts

about movies, triple attributes or graph paths can

narrow down vertex candidates for knowledge se-

lection decision. Moreover, these prior infor-

mation can enhance generalization capability of

knowledge selection models. But it suffers from

information insufﬁciency for response generation

since there is simply a single word or entity to

facilitate generation. For knowledge sentences,

e.g., comments about movies, the text sentences

can provide rich information for generation, but

its unstructured (e.g., document based) representa-

tion scheme demands strong capability for models

to perform knowledge selection or attention from

the list of knowledge texts. Fusion of graph struc-

ture and knowledge sentences m ight yield mutu-

ally reinforcing advantages for knowledge selec-

tion in conversation systems, but there is less study

on that.

To bridge the gap between the two lines of

studies on knowledge aware conversation gener-

ation, we present an augmented knowledge graph

based open-domain chatting m achine (denoted as

AKGCM), which consists of knowledge selec-

tor and response generator. This two-stage ar-

chitecture and graph based knowledge selection

Figure 1: Sample conversation with explainable reasoning

on an augmented knowledge graph.

make our system to be more explainable. Explain-

ability is very important in e.g., information ori-

ented chatting scenarios, where a user needs to

know how new knowledge in chatbot’s responses

is linked to the knowledge in his utterances, or

business scenarios, where a user cannot make a

business decision without justiﬁcation.

To combine graph structure and knowledge sen-

tences, we take a factoid knowledge graph (KG)

as its backbone, and align unstructured sentences

of non-factoid knowledge with the factoid KG by

linking entities from these sentences to vertices

(containing entities) of the K G. Thus we augment

the factoid KG with non-factoid knowledge, and

retain its graph structure. Then we use this aug-

mented KG to facilitate knowledge selection and

response generation, as shown in Figure

For knowledge selection on the graph, we adopt

a deep reinforcement learning (RL) based reason-

ing model(

Das et al., 2018), MINERVA, since it is

as robust as the neural methods due to the use of

continuous vectors for state representation, and is

as explainable as path based symbolic methods be-

cause the agent explicitly traverses the paths in the

graph, and ﬁnally stops at an answer vertex with

correct knowledge. Moreover, our graph differs

from previous KGs in that: some vertices in ours

contain long texts, not a single entity or word. To

fully leverage this long text information, we im-

prove the reasoning algorithm with machine read-

ing comprehension (MRC) technology (

Seo et al.,

2017) to conduct ﬁne-grained semantic matching

between an input message and candidate vertices.

Finally, for response generation, we use an

encoder-decoder model to produce responses con-

ditioned on selected knowledge.

In summary, this paper makes the following

contributions:

• This work is the ﬁrst attempt that uniﬁes

knowledge triples and texts as a graph, and

conducts more ﬂexible multi-hop graph rea-

soning for knowledge selection in conversa-

tion generation. Supported by such knowl-

edge and knowledge selection method, our

system can respond more appropriately and

informatively.

• Our two-stage architecture and graph based

knowledge selection mechanism provide bet-

ter model explainability, which is very impor-

tant for some application scenarios.

• For knowledge selection, to fully leverage

long texts in vertices that differentiate our

graph from others, we improve a state of

the art graph reasoning algorithm (

Das et al.,

2018) with machine reading comprehension

(MRC) technology (

Seo et al., 2017).

2 Related Work

Conversation with Knowledge Graph: There

are growing interests in leveraging factoid knowl-

edge (

Han et al., 2015; Liu et al., 2018; Zhu et al.,

2017) or commonsense knowledge (Young et al.,

2018; Zhou et al., 2018a) with graph based repre-

sentation for generation of appropriate and infor-

mative responses. Compared with them, we aug-

ment previous KGs with knowledge texts and in-

troduce more explainable and ﬂexible multi-hop

graph reasoning models for conversation genera-

tion.

Wu et al. (2018) ever used document reason-

ing network for modeling of conversational con-

texts, but not for knowledge selection. More-

over, previous models cannot effectively utilize

long texts from graph vertices since they simply

use an embedding for representation of the whole

vertex, without further text analysis.

Conversation with Unstructured Texts: With

availability of a large amount of knowledge

texts from Wikipedia or user generated con-

tent websites, e.g., Reddit, some work fo-

cus on either modeling of conversation genera-

tion with unstructured texts (

Ghazvininejad et al.,

2018; Vougiouklis et al., 2016; Xu et al., 2017),

or building benchmark dialogue data grounded

剩余9页未读，继续阅读

评论收藏

内容反馈

Jayxp

粉丝: 6
资源: 137

Knowledge Aware Conversation Generation Reasoning onAugmentedGra...

最新资源

Knowledge Aware Conversation Generation Reasoning onAugmentedGra...

DKN_ Deep knowledge-aware network for news recommendation.pdf

Structure-Aware Human-Action Generation.pdf

藏经阁-RESOURCE AWARE SCHEDULING IN APACHE STORM.pdf

Relation-Aware Pedestrian Attribute Recognition with.pdf

RDMA_Aware_Programming_user_manual.pdf

Cooperation Aware Task Assignment in Spatial Crowdsourcing.pdf

藏经阁-RESOURCE AWARE SCHEDULING IN A.pdf

Ultra Fast Structure-aware Deep Lane Detection.pdf

几何感知的无监督域自适应_Geometry-Aware Unsupervised Domain Adaptation.pdf

Embedding Knowledge Graphs for Semantics-aware Recommendations .pdf

世界5G大会会议报告合集（19份）.zip

2021_Semantic-aware Binary Code Representation with BERT PPT.pdf

藏经阁-Data-Aware Spark.pdf

Morsel-Driven Parallelism A NUMA-Aware Query.pdf

DeepSeek从入门到精通-清华大学-202502.pdf

DeepSeek从入门到精通-清华大学

清华deepseek入门到精通文档 夸克网盘资源下载

DEEP SEEK 本地部署（Ollama + ChatBox）+ 私有知识库（cherry studio）教程

Ollama windows安装包 0.5.7（截止2025-02-01）

人工智能应用：DeepSeek从入门到精通的操作指南与多功能实战详解

YOLOv8-deepsort 实现智能车辆目标检测+车辆跟踪+车辆计数

DeepSeek从入门到精通

浏览器插件Page Assist-1.4.5

AI使用+DeepSeek+DeepSeek清华大学第二版

最新资源

清华deepseek入门到精通文档夸克网盘资源下载