The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
작업 중심 대화 시스템을 위한 기존 파이프라인 방법은 개별적으로 설계되었으며 비용이 많이 듭니다. 기존의 메모리 증강 엔드투엔드 방법은 입력을 출력으로 직접 매핑하고 유망한 결과를 얻습니다. 하지만 기존의 대부분의 엔드투엔드 솔루션은 대화 내역과 지식베이스(KB) 정보를 동일한 메모리에 저장하고 KB 정보를 KB 트리플 형태로 표현하기 때문에 메모리 리더의 메모리 추론을 더욱 어렵게 만들고, 이로 인해 시스템은 응답을 생성하기 위해 메모리에서 올바른 정보를 검색하기가 어렵습니다. 일부 방법에서는 추론을 강화하기 위해 많은 수동 주석을 도입합니다. 수동 주석의 사용을 줄이고 추론을 강화하기 위해 작업 지향 시스템을 위한 계층적 메모리 모델(HM2Seq)을 제안합니다. HM2Seq는 계층적 메모리를 사용하여 대화 기록과 KB 정보를 두 개의 메모리로 분리하고 KB를 KB 행에 저장한 다음 엔터티 디코더와 결합된 메모리 행 포인터를 사용하여 메모리에 대한 계층적 추론을 수행합니다. 공개적으로 사용 가능한 두 가지 작업 지향 대화 데이터 세트에 대한 실험 결과는 우리의 가설을 확인하고 기준선을 능가하여 HM2Seq의 뛰어난 성능을 보여줍니다.
Ya ZENG
Chongqing University
Li WAN
Chongqing University
Qiuhong LUO
Chongqing University
Mao CHEN
Chongqing University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Ya ZENG, Li WAN, Qiuhong LUO, Mao CHEN, "A Hierarchical Memory Model for Task-Oriented Dialogue System" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 8, pp. 1481-1489, August 2022, doi: 10.1587/transinf.2022EDP7001.
Abstract: Traditional pipeline methods for task-oriented dialogue systems are designed individually and expensively. Existing memory augmented end-to-end methods directly map the inputs to outputs and achieve promising results. However, the most existing end-to-end solutions store the dialogue history and knowledge base (KB) information in the same memory and represent KB information in the form of KB triples, making the memory reader's reasoning on the memory more difficult, which makes the system difficult to retrieve the correct information from the memory to generate a response. Some methods introduce many manual annotations to strengthen reasoning. To reduce the use of manual annotations, while strengthening reasoning, we propose a hierarchical memory model (HM2Seq) for task-oriented systems. HM2Seq uses a hierarchical memory to separate the dialogue history and KB information into two memories and stores KB in KB rows, then we use memory rows pointer combined with an entity decoder to perform hierarchical reasoning over memory. The experimental results on two publicly available task-oriented dialogue datasets confirm our hypothesis and show the outstanding performance of our HM2Seq by outperforming the baselines.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022EDP7001/_p
부
@ARTICLE{e105-d_8_1481,
author={Ya ZENG, Li WAN, Qiuhong LUO, Mao CHEN, },
journal={IEICE TRANSACTIONS on Information},
title={A Hierarchical Memory Model for Task-Oriented Dialogue System},
year={2022},
volume={E105-D},
number={8},
pages={1481-1489},
abstract={Traditional pipeline methods for task-oriented dialogue systems are designed individually and expensively. Existing memory augmented end-to-end methods directly map the inputs to outputs and achieve promising results. However, the most existing end-to-end solutions store the dialogue history and knowledge base (KB) information in the same memory and represent KB information in the form of KB triples, making the memory reader's reasoning on the memory more difficult, which makes the system difficult to retrieve the correct information from the memory to generate a response. Some methods introduce many manual annotations to strengthen reasoning. To reduce the use of manual annotations, while strengthening reasoning, we propose a hierarchical memory model (HM2Seq) for task-oriented systems. HM2Seq uses a hierarchical memory to separate the dialogue history and KB information into two memories and stores KB in KB rows, then we use memory rows pointer combined with an entity decoder to perform hierarchical reasoning over memory. The experimental results on two publicly available task-oriented dialogue datasets confirm our hypothesis and show the outstanding performance of our HM2Seq by outperforming the baselines.},
keywords={},
doi={10.1587/transinf.2022EDP7001},
ISSN={1745-1361},
month={August},}
부
TY - JOUR
TI - A Hierarchical Memory Model for Task-Oriented Dialogue System
T2 - IEICE TRANSACTIONS on Information
SP - 1481
EP - 1489
AU - Ya ZENG
AU - Li WAN
AU - Qiuhong LUO
AU - Mao CHEN
PY - 2022
DO - 10.1587/transinf.2022EDP7001
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2022
AB - Traditional pipeline methods for task-oriented dialogue systems are designed individually and expensively. Existing memory augmented end-to-end methods directly map the inputs to outputs and achieve promising results. However, the most existing end-to-end solutions store the dialogue history and knowledge base (KB) information in the same memory and represent KB information in the form of KB triples, making the memory reader's reasoning on the memory more difficult, which makes the system difficult to retrieve the correct information from the memory to generate a response. Some methods introduce many manual annotations to strengthen reasoning. To reduce the use of manual annotations, while strengthening reasoning, we propose a hierarchical memory model (HM2Seq) for task-oriented systems. HM2Seq uses a hierarchical memory to separate the dialogue history and KB information into two memories and stores KB in KB rows, then we use memory rows pointer combined with an entity decoder to perform hierarchical reasoning over memory. The experimental results on two publicly available task-oriented dialogue datasets confirm our hypothesis and show the outstanding performance of our HM2Seq by outperforming the baselines.
ER -