Img Img


Dialog System Technology Challenges
Honolulu, Hawaii, USA, January 27, 2019

Workshop Program

09:00-09:10Welcome Remarks from DSTC Organizing Committee (Paper) (Slides)Koichiro Yoshino
09:10-09:25DSTC7 Task 1: Noetic End-to-End Response Selection - Track 1 Overview Chulaka Gunasekara, Jonathan K. Kummerfeld, Lazaros Polymenakos, and Walter S. Lasecki
09:25-09:40Grounded Response Generation Task at DSTC7 (Slides) - Track 2 OverviewMichael Galley, Chris Brockett, Xiang Gao, Jianfeng Gao, and Bill Dolan
09:40-09:55Audio Visual Scene-aware dialog (AVSD) Track for Natural Language Generation in DSTC7 - Track 3 OverviewHuda Alamari, Chiori Hori, Tim K. Marks, Dhruv Batra, and Devi Parikh
Track 1 Oral Session: Sentence Selection
09:55-10:15Sequential Attention-based Network for Noetic End-to-End Response SelectionQian Chen and Wen Wang (Slides)
10:15-10:35Comparison of Transfer-Learning Approaches for Response Selection in Multi-Turn ConversationsJesse Vig and Kaiai Ramea (Slides)
10:35-10:55Knowledge-incorporating ESIM models for Response Selection in Retrieval-based Dialog SystemsJatin Ganhotra, Siva Sankalp Patel and Kshitij Fadnis
10:55-11:10Coffee break
Invited Talks
11:10-12:10 Massively Multilingual Dialog and Q&A Holger Schwenk (Facebook AI Research, Paris)
12:10-12:15Sponsor TalkWeChat
12:15-12:20Sponsor TalkAdobe Research
12:20-12:25Sponsor TalkUber AI
12:25-13:25Poster session 1 w. lunch
 Building Sequential Inference Models for End-to-End Response SelectionJia-Chen Gu, Zhen-Hua Ling, Yuping Ruan and Quan Liu
 RAP-Net: Recurrent Attention Pooling Networks for Dialogue Response SelectionChao-Wei Huang, Ting-Rui Chiang, Shang-Yu Su and Yun-Nung Chen
 End-to-end Gated Self-attentive Memory Network for Dialog Response SelectionShuo Sun, Yik-Cheung Tam, Jie Cao, Can-Xiang Yan, Zuohui Fu, Cheng Niu and Jie Zhou
 Enhanced Sequential Representation Augmented with Utterance-level Attention for Response SelectionTaesun Whang, Dongyub Lee, Chanhee Lee and Heuiseok Lim
 Multi-level Context Response Matching in Retrieval-Based Dialog SystemsBasma El Amel Boussaha, Nicolas Hernandez, Christine Jacquin and Emmanuel Morin
 Spatio-Temporal Matching Network for Multi-Turn Responses Selection in Retrieval-Based ChatbotsJunyu Lu, Zeying Xie, Guang Ling, Chao Zhou and Zenglin Xu
 End-to-End Multimodal Dialog Systems with Hierarchical Multimodal Attention on Video FeaturesHung Le, Steven C.H. Hoi, Doyen Sahoo and Nancy F. Chen
 Reactive Multi-Stage Feature Fusion for Multimodal Dialogue ModelingYi-Ting Yeh, Tzu-Chuan Lin, Hsiao-Hua Cheng, Yu-Hsuan Deng, Shang-Yu Su and Yun-Nung Chen
 Entropy-Enhanced Multimodal Attention Model for Scene-Aware Dialogue GenerationKuan-Yen Lin, Chao-Chun Hsu, Yun-Nung Chen and Lun-Wei Ku
13:25-14:25Poster session 2 w. lunch
 End-to-End Question Answering Models for Goal-Oriented Dialog LearningJamin Shin, Andrea Madotto, Minjoon Seo and Pascale Fung
 Noetic End-to-End Response Selection with Supervised Neural Network Based Classifiers and Unsupervised Similarity ModelsPawel Skórzewski, Weronika Sieińska and Marek Kubis
 Convolutional Neural Encoder for the 7th Dialogue System Technology ChallengeMandy Korpusik and James Glass (Poster)
 Learning Multi-Level Information for Dialogue Response Selection by Highway Recurrent TransformerTing-Rui Chiang, Chao-Wei Huang, Shang-Yu Su and Yun-Nung Chen
 Top-K Attention Mechanism for Complex Dialogue SystemChang-Uk Shin and Jeong-Won Cha (Slides)
 Promoting Diversity for End-to-End Conversation Response GenerationYu-Ping Ruan, Zhen-Hua Ling, Quan Liu, Jia-Chen Gu and Xiaodan Zhu
 Knowledge-Grounded Response Generation with Deep Attentional Latent-Variable ModelHao-Tong Ye, Kai-Ling Lo, Shang-Yu Su and Yun-Nung Chen
 Investigation of Attention-Based Multimodal Fusion and Maximum Mutual Information Objective for DSTC7 Track3Bairong Zhuang, Wenbo Wang and Takahiro Shinozaki (Poster)
 Context, Attention and Audio Feature Explorations for Audio Visual Scene-Aware DialogShachi H Kumar, Eda Okur, Saurav Sahay, Juan Jose Alvarado Leanos, Jonathan Huang and Lama Nachman
 Neural Machine Translation For Paraphrase GenerationAlex Sokolov
 Dynamic Memory Networks for Dialogue Topic TrackingSeokhwan Kim
Track 2 Oral Session: Sentence Generation
14:25-14:45Cluster-based Beam Search for Pointer-Generator Chatbot Grounded by KnowledgeYik Cheung Tam, Jiachen Ding, Cheng Niu and Jie Zhou (Slides)
14:45-15:05An Ensemble Dialogue System for Facts-Based Sentence GenerationRyota Tanaka, Akihide Ozeki, Shugo Kato and Akinobu Lee (Slides)
15:05-15:25The OneConn-MemNN System for Knowledge-Grounded Conversation ModelingJunyuan Zheng, Surya Kasturi, Mason Lin, Xin Chen, Onkar Salvi, and Harry Jiannan Wang
15:25-15:40Coffee break
Track 3 Oral Session: Audio Visual Scene Aware Dialog
15:40-16:00From FiLM to Video: Multi-turn Question Answering with Multi-modal ContextDat Nguyen, Shikhar Sharma, Hannes Schulz and Layla El Asri
16:00-16:20DSTC7-AVSD: Scene-Aware Video-Dialogue Systems with Dual AttentionRamakanth Pasunuru and Mohit Bansal
16:20-16:40CMU Sinbad's Submission for the DSTC7 AVSD ChallengeRamon Sanabria, Shruti Palaskar and Florian Metze (Slides)
Future DSTC & Closing
16:40-17:50DSTC8 Track ProposalsSeokhwan Kim
16:40-16:50End-to-end Task Completion Dialog ChallengeMicrosoft Research & Tsinghua University
16:50-17:00Towards Fast Adaptation of User Simulation to New DomainsMicrosoft Research in Montreal
17:00-17:10Multi-Domain Human-to-Human Dialog Understanding, a machine reading approachNaver Labs Europe & Adobe Research
17:10-17:20Dialog State Tracking for Conversational Image EditingAdobe Research
17:20-17:30NOESIS II: Predicting Responses, Identifying Success, and Managing Complexity in Task-Oriented DialogueIBM & University of Michigan
17:30-17:40Audio Visual Scene-Aware Dialog TrackMERL & CMU
17:40-17:50Scalable Schema-Guided Dialogue State TrackingGoogle
17:50-18:00Wrap-up TalkChiori Hori

Workshop Information

DSTC7 Workshop will be held at AAAI on January 27, 2019 at the Hilton Hawaiian Village, Honolulu, Hawaii, USA. More information about registration will be provided later.

Keynote speaker

Title: Massively Multilingual Dialog and Q&A

Speaker: Holger Schwenk, Facebook AI Research, Paris
Time: 10:50-11:50

Abstract: Dialog systems and Q&A approaches focus mainly on the English language. Coverage of other languages is limited to high-resource languages and truly multilingual systems which can transparently switch languages are missing. We first show how to learn massively multilingual embeddings of sentences and paragraphs. We only need publicly available resources to cover more than ninety languages in many different scripts. We then show how this universal representations can used as building block in dialog and Q&A systems. The talk concludes with initial experimental results.

Important Dates

Paper submission deadline 5 Nov 2018 (Extended to 16 Nov 2018, 23:59 Pacific Time)
Paper acceptance notification 26 Nov 2018 (Extended to 28 Nov 2018)
Camera-ready papers deadline 19 Dec 2018
DSTC7 workshop @ AAAI 27 January 2019

Summary paper: Bibtex reference

We will prepare the challenge summary paper as follows.

title={The 7th dialog system technology challenge},
author={Koichiro Yoshino and Chiori Hori and Julien Perez and Luis Fernando D'Haro and Lazaros Polymenakos and Chulaka Gunasekara and Walter S. Lasecki and Jonathan Kummerfeld and Michael Galley and Chris Brockett and Jianfeng Gao and Bill Dolan and Sean Gao and Tim K. Marks and Devi Parikh and Dhruv Batra},
journal={arXiv preprint},

Paper Format

Submissions should be made via Microsoft-CMT and must follow the formatting guidelines for AAAI-2019 (use the AAAI Author Kit). All submissions must be anonymous and conform to AAAI standards for double-blind review. The papers adhere to the 2-column AAAI format in 7 page long, 6 pages for the contents and remaining 1 page only for the reference, will be considered for review.

Author Kit:


Submit you paper through Submission System

Workshop Sponsors