GMT/UTCTime(local)SessionDetails
DAY 1: November 15
07:00-08:3015:00-16:30Steering committee meeting
12:00-12:3020:00-20:30Opening
12:30-13:3020:30-21:30Keynote 1
(live presentation)
Maxine Eskenazi, Carnegie Mellon University
Title: User-centric dialog
Abstract: Some recent research has turned from being agent-centric to being user-centric. This paradigm shift is important if we are to create systems acceptable to the general population of users. In this talk we will begin with the reasoning behind user-centric research. Then we will look at concrete ways to apply this point of view to system training and to evaluation. Then we will address user-centric strategies when dealing with a malevolent user.
13:30-13:4021:30-21:40Break
13:40-13:5221:40-21:52Natural Language Understanding Out-of-Scope Domain and Intent Classification through Hierarchical Joint Modeling,
Pengfei Liu, Kun Li and Helen Meng
13:52-14:0421:52-22:04Segmentation-Based Formulation of Slot Filling Task for Better Generative Modeling,
Kei Wakabayashi, Johane Takeuchi and Mikio Nakano
14:04-14:1622:04-22:16Active Learning and Multi-label Classification for Ellipsis and Coreference Detection in Conversational Question-Answering,
Quentin Brabant, Lina Maria Rojas-Barahona and Claire Gardent
14:16-14:2822:16-22:28Can we predict how challenging Spoken Language Understanding corpora are across sources, languages and domains?
Frederic Bechet, Christian Raymond, Achraf Hamane, Rim Abrougui, Gabriel Marzinotto and Géraldine Damnati
14:30-14:4022:30-22:40Break
14:40-14:5222:40-22:52Personalisation and GenerationPersonalized Extractive Summarization with Discourse Structure Constraints Towards Efficient and Coherent Dialog-based News Delivery,
Hiroaki Takatsu, Ryota Ando, Hiroshi Honda, Yoichi Matsuyama and Tetsunori Kobayashi
14:52-15:0422:52-23:04Empathetic Dialogue Generation with Pre-trained RoBERTa-GPT2 and External Knowledge,
Ye Liu, Wolfgang Maier, Wolfgang Minker and Stefan Ultes
15:04-15:1623:04-23:16Towards Handling Unconstrained User Preferences,
Suraj Pandey, Svetlana Stoyanchev and Rama Doddipatla
15:16-15:2823:16-23:28Jurassic is (almost) All You Need: Few-Shot Meaning-to-Text Generation for Open-Domain Dialogue,
Lena Reed, Cecilia Li, Angela Ramirez, Liren Wu and Marilyn Walker
15:30-16:0023:40-00:10Coffee break / Informal discussion
DAY 2: November 16
11:50-12:2019:50-20:20Coffee break / Informal discussion
12:30-12:4220:30-20:42Spoken and Multimodal InteractionComparison of Automatic Speech Recognition Systems,
Joshua Kim, Chunfeng Liu, Rafael A. Calvo, Kathryn McCabe, Silas C.R. Taylor, Björn W. Schuller and Wu Kaihang
12:42-12:5420:42-20:54Multimodal Dialogue Response Timing Estimation Using Dialogue Context Encoder,
Ryota Yahagi, Yuya Chiba, Takashi Nose and Akinori Ito
12:54-13:0620:54-21:06Eliciting Cooperative Persuasive Dialogue by Multimodal Emotional Robot,
Sara Asai, Koichiro Yoshino, Seitaro Shinagawa, Sakriani Sakti and Satoshi Nakamura
13:06-13:1821:06-21:18Voice Activity Detection Considerations in a Dialog Agent for Dysarthric Speakers,
Jackson Liscombe, Hardik Kothare, Michael Neumann, Oliver Roesler, Doug Habberstad, Andrew Cornish, David Pautler, David Suendermann-Oeft, Vikram Ramanarayanan and Alexander Ocampo
13:30-14:0021:30-22:00Sponsor Sessions
14:15-15:1522:15-23:15Keynote 2Helen Hastie, Heriot Watt University
Title: Trustworthy Interactive Robots
Abstract: Trust is a multifaceted, complex phenomenon that is not well understood when it occurs between humans, let alone between humans and robots. Robots that portray social cues, including voice, gestures and facial expressions, are key tools in researching human-robot trust, specifically how trust is established, lost and regained. In this talk, I will discuss various aspects of trust for HRI including language, social cues, embodiment, transparency, mental models and theory of mind. I will present a number of studies performed in the context of two large projects: the UKRI Trustworthy Autonomous Systems Programme, specifically the Node on Trust; and the EPSRC ORCA Hub for robotic and autonomous systems for remote hazardous environments. This work will be contextualised around the new National Robotarium opening soon in Edinburgh.
15:15-15:2023:15-23:20Break
15:20-15:3223:20-23:32EvaluationDesign guidelines for developing systems for dialogue system competitions,
Kazunori Komatani, Ryu Takeda, Keisuke Nakashima, Mikio Nakano
15:32-15:4423:32-23:44Modeling Performance in Open-Domain Dialogue with PARADISE,
Marilyn Walker, Stephen Whittaker, Colin Harmon and James Graupera
15:44-15:5623:44-23:56Understanding How People Rate Their Conversations,
Alexandros Papangelis, Nicole Chartier, Pankaj Rajan, Julia Hirschberg and Dilek Hakkani-Tur
16:00-16:3000:00-00:30Coffee break / Informal discussion
DAY 3: November 17
10:30-11:3018:30-19:30Singapore Virtual Tour
11:40-12:1019:40-20:10Coffee break / Informal discussion
12:30-12:4220:30-20:42Dialogue Modelling and ApplicationsA WoZ Study for an Incremental Proficiency Scoring Interview Agent Eliciting Ratable Samples,
Mao Saeki, Weronika Demkow, Tetsunori Kobayashi and Yoichi Matsuyama
12:42-12:5420:42-20:54SUPPLE: A Dialogue Management Approach based on Conversation Patterns,
Florian Kunneman and Koen Hindriks
12:54-13:0620:54-21:06Dialogue Management as Graph Transformations,
Nicholas Walker, Torbjørn Dahl and Pierre Lison
13:06-13:2021:06-21:20Break
13:20-14:0021:20-22:00WOCHAT
Special Session 1/2
Data collection for detecting the unwillingness to answer questions in dialogue
Kazumi Nagao, Ryuichiro Higashinaka and Kazuto Ataka
Enhancing Self-Disclosure In Neural Conversation Models By Response Candidate Re-ranking
Mayank Soni, Benjamin Cowan and Vincent Wade
On the Impact of Self-efficacy on Assessment of User Experience in Customer Service Chatbot Conversations
Yuexin Cao, Vicente Ivan Sanchez Carmona, Xiaoyi Liu, Changjian Hu, Neslihan Iskender, André Beyer, Sebastian Möller and Tim Polzehl
14:00-14:0522:00-22:05Break
14:10-14:5022:10-22:50WOCHAT
Special Session 2/2
Learning to ask specific questions naturally in chat-oriented dialogue systems,
Sota Horiuchi and Ryuichiro Higashinaka
Fine-tuning a pre-trained Transformer-based encoder-decoder model with user-generated question-answer pairs to realize character-like chatbots,
Koh Mitsuda, Ryuichiro Higashinaka, Hiroaki Sugiyama, Masahiro Mizukami, Tetsuya Kinebuchi, Ryuta Nakamura, Noritake Adachi and Hidetoshi Kawabata
Investigating the Impact of Pre-trained Language Models on Dialog Evaluation,
Chen Zhang, Luis Fernando D’haro, Thomas Friedrichs, Haizhou Li and Yiming Chen
14:50-15:0022:50-23:00Break
15:00-16:0023:00-00:00Keynote 3Jinho D. Choi, Emory University
Title: Alexa Prize and Beyond: the Future of Chatbot
Abstract: Developing a robust dialogue system for open-domain conversations is challenging because it is difficult to collect large data to train deep learning models that cover a variety of topics and there is no “ground truth” way of conducting open-domain conversations that satisfies a wide range of people. Even the evaluation of dialogue management is often subjective (thus, bias), which adds another level of difficulty to enhance open-domain dialogue systems. In this talk, I will first illustrate limitations of state-of-the-art dialogue systems using the latest transformer models as well as top-ranked bots from the Alexa Prize Socialbot Grand Challenge. I will then introduce our inference-driven dialogue management framework and discuss its extension to deep learning-based dialogue models. Finally, I will present real-life applications of open-domain dialogue management that we are currently working on regarding education and healthcare.
16:00-16:3000:00-00:30Closing