Call for Participation
CALL
----------------------------------------------------------------------
We are sending the second call for participation in The 10th dialogue system technology challenge. The challenge periods are fixed, and now we have opened the registration form! We look forward to your participation.
===========================================================================================
The Tenth Dialog System Technology Challenge (DSTC10)
Call for Participation
Website: https://sites.google.com/dstc.community/dstc10/
===========================================================================================
Background
The DSTC shared tasks have provided common testbeds for the dialog research community since 2013. From its sixth edition, it has been rebranded as "Dialog System Technology Challenge" to cover a wider variety of dialog-related problems.
For this year's challenge, we opened the call for track proposals and selected the following five parallel tracks by peer-reviews:
- MOD: Internet Meme Incorporated Open-domain Dialog
Training data release: 6/14 Test data release: 9/13
https://openai.weixin.qq.com/dstc/DescriptionZH
- Knowledge-grounded Task-oriented Dialogue Modeling on Spoken Conversations
Validation data release: 6/14 Test data release: 9/13
https://github.com/alexa/alexa-with-dstc10-track2-dataset
- SIMMC 2.0: Situated Interactive Multimodal Conversational AI
Training data release: 6/14 Test data release: 9/13
https://github.com/facebookresearch/simmc2
- Reasoning for Audio Visual Scene-Aware Dialog
Training data release: 6/14 Test data release: 9/13
https://github.com/dialogtekgeek/AVSD-DSTC10_Official
- Automatic Evaluation and Moderation of Open-domain Dialogue Systems
Validation data release - Subtask 1: 6/14 Training data release - Subtask 2: 6/14 Test data release: 9/13
Participation is welcomed from any research team (academic, corporate, non-profit, government).
Registration:
Please fill the form with your name, affiliation and contact information from the following google form. Track organizers will contact you at the training data release phase.
https://forms.gle/Qigb3N3hGqpEgsuW8
IMPORTANT DATES
---------------
* Training data release: Jun 14th
* Test data release: Sep 13th ; 12 weeks development
* System submission starts (evaluation track): Sep 13th
* Submission of final results (any tracks): Sep 21st
* Final result announcement: Oct 1st - Oct 8th
* Paper submission: Nov 1st
* Workshop: January or February (Venue: TBD)
FAQ
Q: Why we need pre-registration?
A: We send you the detailed information of your registered track after the training release. Please check the tracks you are interested in.
Q: How do we update the information if we want to change the track to participate?
A: Please put your new interests with the same e-mail address. We will send you the information according to your latest information.
Q: Is it necessary to participate in the challenge if we make pre-registration?
A: You can withdraw your entry before the test data release. We ask you again before the test data release whether you will submit the final results or not. After you receive the test data, we request you to submit at least one entry as your result.
Q: Would the data/resources be released only to the registered participants?
A: We will notify the latest information to people who pre-registered. Finally, the data will be opened, but during the challenge period, each track organizer will decide whether they will make open the data or not.
ORGANIZING COMMITTEE
--------------------
General chair: Koichiro Yoshino (RIKEN)
Workshop chairs: Yun-Nung (Vivian) Chen (NTU), Paul Crook (Facebook)
Publication chairs: Satwik Kottur (Facebook), Jinchao Li (Microsoft)
Publicity chairs: Behnam Hedayatnia (Amazon), Seungwhan (Shane) Moon (Facebook)
DSTC10 Track Organizers
--------------------
MOD: Internet Meme Incorporated Open-domain Dialog
- Zhengcong Fei, Zekang Li (Chinese Academy of Sciences)
- Jinchao Zhang, Jie Zhou (WeChat AI)
- Yang Feng (Institute of Computing Technology)
Knowledge-grounded Task-oriented Dialogue Modeling on Spoken Conversations
- Seokhwan Kim, Yang Liu, Di Jin, Alexandros Papangelis, Behnam Hedayatnia, Karthik Gopalakrishnan, Dilek Hakkani-Tur (Amazon)
SIMMC 2.0: Situated Interactive Multimodal Conversational AI
- Seungwhan Moon, Satwik Kottur, Paul A. Crook, Ahmad Beirami, Eunjoon Cho, Babak Damavandi, Alborz Geramifard, Rajen Subba (Facebook)
Reasoning for Audio Visual Scene-Aware Dialog
- Ankit Shah (Carnegie Mellon University)
- Shijie Geng (Rutgers University)
- Peng Gao (The Chinese University of Hong Kong)
- Anoop Cherian, Chiori Hori, Tim K. Marks (Mitsubishi Electric Research Laboratories)
Automatic Evaluation and Moderation of Open-domain Dialogue Systems
- Zhang Chen(National University of Singapore)
- João Sedoc (New York University)
- Luis Fernando D’Haro (Universidad Politécnica de Madrid)
- Rafael Banchs, Alexander Rudnicky (Carnegie Mellon University)
DSTC Steering Committee
---------------------------------------
- Koichiro Yoshino - Nara Institute of Science and Technology (NAIST), Japan
- Chiori Hori - Mitsubishi Electric Research Laboratories (MERL), USA
- Rafael Banchs - Carnegie Mellon University, USA
- Micheal Galley - Microsoft Research AI, USA
CONTACT
-------
Please feel free to post any question or comment to the DSTC mailing list list@dstc.community.
To join the mailing list: visit https://groups.google.com/a/dstc.community/forum/#!forum/list/join.