DSTC8 Workshop @ AAAI-20

Workshop Information

DSTC8 Workshop will be held at AAAI-20 on February 8, 2020 at the Hilton New York Midtown, New York, USA.

The AAAI-20 online registration is available at https://aaaiconf.cvent.com/AAAI-20.

Workshop Program

09:20-09:30 Opening Welcome Remarks, Seokhwan Kim [slides]

Track 1: Multi-domain Task Completion

(Chair: Jinchao Li & Hannes Schulz)

09:30-09:50 Results of the Multi-Domain Task-Completion Dialog Challenge, Jinchao Li, Hannes Schulz, Adam Atkinson, Mahmoud Adada, Baolin Peng, Jianfeng Gao, Sungjin Lee, Minlie Huang, Qi Zhu, Ryuichi Takanobu

09:50-10:10 End-to-end neural architecture for pipelined dialogue system using GPT-2, Donghoon Ham*, Jeong-Gwan Lee*, Youngsoo Jang, Kee-Eung Kim (* Student Travel Grant Winners)

10:10-10:30 Hybrid Generative-Retrieval Transformers for Fast Domain Adaptation, Igor Shalyminov, Alessandro Sordoni, Adam Atkinson, Hannes Schulz

10:30-10:50 Coffee Break

Track 2: NOESIS II - Predicting Responses, Identifying Success, and Managing Complexity in Task-Oriented Dialogue

(Chair: Jonathan Kummerfeld)

10:50-11:10 NOESIS II: Predicting Responses, Identifying Success, and Managing Complexity in Task-Oriented Dialogue, Chulaka Gunasekara, Jonathan K Kummerfeld, Luis Lastras, Walter Lasecki

11:10-11:30 Transformer-based Semantic Matching Model for Neotic Response Selection, Yiying Yang, Kaijie Zhou, Xuan Li, Dawei Zhu, Xiaoyuan Yao, Jianping Shen

11:30-11:50 Enhancing Response Selection with Advanced Context Modeling and Post-training, ShuangZhi Wu, Yufan Jiang, Xu Wang, Wei Miao, Zhenyu Zhao, Xie Jun, Mu Li

Track 1 & 2

11:50-12:40 Poster Session (Chair: Jinchao Li, Hannes Schulz & Jonathan Kummerfeld)

• A BERT-based dialogue Agent for Multi-Domain Task-Completion dialogue Challenge, Haoqi Xiong, Miao Li, Yunbo Cao
• Hierarchical Context Enhanced Multi-Domain Dialogue System, Jingyuan Yang, Jingyuan Yang, Liu Guang, Mao Yuzhao, Zhao Zhiwei, Gao Weiguo, Li Xuan, Jianping Shen
• Show Us the Way: Learning to Manage Dialog from Demonstrations, Gabriel Gordon-Hall, Philip John Gorinski, Gerasimos Lampouras, Ignacio Iacobacci
• Dropout-based Bayesian Q-network for Task-oriented Dialog Systems, Chansong Jo, Jeonpil Lee, Myoung-Wan Koo
• Multi-domain Task Oriented Dialogue System Using Word Constraints, Chang-Uk Shin, Jeong-Won Cha
• Attention based E2E system for next utterance selection of DSTC8, Tae-hoon Kim, Do-hyeon Kim, Seung-hyuk Choi
• Pre-Trained and Attention-Based Neural Networks for Building Noetic Task-Oriented Dialogue Systems, Jia-Chen Gu, Tianda Li, Quan Liu, Xiaodan Zhu, Zhen-Hua Ling, Yu-Ping Ruan
• Model Ensembling of ESIM and BERT for Dialogue Response Selection, Dario Bertero, Takeshi Homma, Kenichi Yokote, Makoto Iwayama, Kenji Nagamatsu
• Leveraging Speaker Profile and Knowledge Enrichment for End-to-End Advising Response Ranking, Sung-Ping Chang, Yun-Nung Chen, Ting-Rui Chiang, Chao-Wei Huang
• Speaker-Aware BERT for Multi-Party Dialog Response Selection, Tatsuya Nishiyama, Ryota Tanaka, Yuya Ishijima, Akinobu Lee
• A randomly wired network based on RoBERTa for sentence classification, Byoungjae Kim, Jungyun Seo, Myoung-Wan Koo
• Reinforcement Learning of Multi-Domain Dialog Policies Via Action Embeddings, Jorge A Mendez, Alborz Geramifard, Mohammad Ghavamzadeh, Bing Liu

12:40-13:40 Lunch Break

Track 3: Audio Visual Scene-aware Dialog

(Chair: Tim K Marks)

13:40-14:00 Audio Visual Scene-Aware Dialog (AVSD) Track for Natural Language Generation in DSTC8, Chiori Hori, Anoop Cherian, Takaaki Hori, Tim K. Marks

14:00-14:20 Bridging Text and Video: A Universal Multimodal Transformer for Video-Audio Scene-Aware Dialog, Zekang Li, Zongjia Li, Jinchao Zhang, Yang Feng, Cheng Niu, Jie Zhou

14:20-14:40 Multi-step Joint-Modality Attention Network for Scene-Aware Dialogue System, Yun-Wei Chu, Kuan-Yen Lin, Chao-Chun Hsu, Lun-Wei Ku

Track 4: Schema-Guided Dialogue State Tracking

(Chair: Srinivas Sunkara & Xiaoxue Zang)

14:40-15:00 Schema-Guided Dialogue State Tracking Task at DSTC8, Abhinav Rastogi, Xiaoxue Zang, Srinivas Sunkara, Raghav Gupta, Pranav Khaitan

15:00-15:20 An End-to-End Dialogue State Tracking System with Machine Reading Comprehension and Wide & Deep Classification, Yue Ma, Zengfeng Zeng, Dawei Dawei Zhu, Xuan Li, Yiying Yang, Xiaoyuan Yao, Kaijie Zhou, Jianping Shen

15:20-15:40 The SPPD System for Schema Guided Dialogue State Tracking Challenge, Miao Li, Haoqi Xiong, Yunbo Cao

15:40-16:00 Coffee Break

Track 3 & 4

16:00-16:50 Poster Session (Chair: Tim K Marks, Srinivas Sunkara & Xiaoxue Zang)

• Audio Visual Scene-Aware Dialog System Using Dynamic Memory Networks, Huiyuan Xie, Ignacio Iacobacci
• Multimodal Transformer with Pointer Network for the DSTC8 AVSD Challenge, Hung Le, Nancy Chen
• Pre-training Assisted Scene-aware Dialogue Generation, Junkun Chen, Huang He, Siqi Bao, Fan Wang
• DSTC8-AVSD: Multimodal Semantic Transformer Network with Retrieval StyleWord Generator, Hwanhee Lee, Seunghyun Yoon, Franck Dernoncourt, Doo Soon Kim, Trung Bui, Kyomin Jung
• Spatio-Temporal Scene Graph Reasoning for Audio Visual Scene-Aware Dialog at DSTC8, Shijie Geng, Gao Peng, Chiori Hori, Anoop Cherian
• A BERT-based Unified Span Detection Framework for Schema-Guided Dialogue State Tracking, Xing Shi, Scot Fang, Kevin Knight
• Zero-Shot State Tracking and User Adoption Tracking on Schema-Guided Dialogue, Shuyu Lei, Shuaipeng Liu, Mengjun Shen, Huixing Jiang, Xiaojie Wang
• Fine-Tuning BERT for Schema-Guided Zero-Shot Dialogue State Tracking, Yu-Ping Ruan, Zhen-Hua Ling, Jia-Chen Gu, Quan Liu
• LION-Net: LIghtweight ONtology-independent Networks for Schema-Guided Dialogue State Generation, Kai-Ling Lo, Ting-Wei Lu, Tzu-teng Weng, Yun-Nung Chen, I-Hsuan Chen
• Goal-Oriented Multi-Task BERT-based Dialogue State Tracker, Pavel Gulyaev, Evgenia Elistratova, Vasily Konovalov, Yury Kuratov, Leonid P. Pugachev, Mikhail Burtsev
• Domain-Aware Dialogue State Tracker For Multi-Domain Dialogue Systems, Vevake Balaraman, Bernardo Magnini
• Candidate Attended Dialogue State Tracking Using BERT, Junyuan Zheng, Onkar Salvi, John Chan
• Multimodal Dialogue State Tracking By QA Approach with Data Augmentation, Xiangyang Mou, Brandyn Sigouin, Ian Steenstra, Hui Su

Future DSTC & Closing

16:50-17:50 DSTC9 Track Proposals (Chair: Abhinav Rastogi)

2 mins oral presentation + 40 mins poster presentation

• Common Grounding Dialog Challenge, The University of Tokyo, The Graduate University for Advanced Studies, & National Institute of Informatics
• Towards Compositional Representations for Interactive Language Learning: CompGuessWhat?!, Heriot-Watt University, University of Copenhagen, University of Edinburgh, & Edinburgh Centre for Robotics
• Beyond Domain APIs: Task-oriented Conversational Modeling with Unstructured Knowledge Access, Amazon Alexa AI
• Contextual Task-Oriented Language Generation, Uber AI
• Multi-domain Task-oriented Dialog Challenge II, Microsoft Research AI & Tsinghua University
• Interactive Evaluation of Dialog, CMU & USC
• Non-collaborative Dialog System Challenge, University of California, Davis
• SIMMC: Situated Interactive Multi-Modal Conversational AI, Facebook Assistant & Facebook AI
• Doc2Dial: Document Grounded Dialogue Modeling, IBM Research AI

17:50-18:00 Wrap-up Talk, Michel Galley [slides]

Important Dates

• Paper submission deadline: Nov 17, 2019
• Paper acceptance notification: Dec 3, 2019
• AAAI early registration deadline: Dec 13, 2019
• Camera-ready submission deadline: Dec 18, 2019
• AAAI online registration deadline: Jan 10, 2020
• DSTC8 workshop @ AAAI: Feb 8, 2020

Workshop Sponsors