DSTC11 @ INLG-SIGdial 2023
Workshop Program
Full Program – September 11, 2023 (Prague time / CEST / GMT+2)
Presentation format:
Talks (in-person and virtual): each talk is 20 minutes including Q&A (remote presenters will join via Zoom.)
Physical posters: poster boards at the venue are A0 portrait size.
Virtual posters: hosted on a Discord server. Each poster will have its own Discord channel with a pdf of the poster to upload and discussions of the poster in text chat. We encourage authors of poster papers to check their channel occasionally for the duration of DSTC11 and SIGdial and respond to questions. Optionally, authors of poster papers can also present their poster live at 17:40-18:40 local time on September 11 (using video and screen sharing). Note: authors of physical posters are welcome to use the Discord server to present their work to remote attendees.
Accessing Zoom and Discord links for virtual presentation: to access links to Zoom (talks) and Discord (virtual posters), please visit the conference's online participation page. As this page prompts you to create an account, please use the email address associated with your Eventbrite registration.
Please email Michel Galley (mgalley@acm.org) for any questions.
Opening
09:00 - 09:15 Opening Remarks [slides]
Track 1 session: Ambiguous Candidate Identification and Coreference Resolution for Immersive Multimodal Conversations
Session Chair: Satwik Kottur
09:15 - 09:35 Track 1 Overview
Overview of Situated and Interactive Multimodal Conversations (SIMMC) 2.1 Track at DSTC 11
Kottur, Satwik; Moon, Seungwhan
09:35 - 09:55 Best Paper
Improving Situated Conversational Agents with Step-by-Step Multi-modal Logic Reasoning
Long, Yuxing; Zhang, Huibin; Hui, Binyuan; Yang, Zhenglu; Yuan, Caixia; Wang, Xiaojie; Huang, Fei; Li, Yongbin
Track 2 talks: Intent Induction from Conversations for Task-Oriented Dialogue
Session Chair: Raphael Shu
09:55 - 10:15 Track 2 Overview
Intent Induction from Conversations for Task-Oriented Dialogue Track at DSTC 11
Gung, James; Shu, Raphael; Moeng, Emily; Rose, Wesley; Romeo, Salvatore; Gupta, Arshit; Benajiba, Yassine; Mansour, Saab; Zhang, Yi
10:15 - 10:35 Best Paper
A Two-Stage Progressive Intent Clustering for Task-Oriented Dialogue
Du, Bingzhu; Su, Nan; Zhang, Yuchi; Wang, Yongliang
10:35 - 11:00 Coffee break
Track 3 talks: Speech-Aware Dialog Systems Technology Challenge
Session Chair: Raghav Gupta
11:00 - 11:20 Track 3 Overview
DSTC-11: Speech Aware Task-Oriented Dialog Modeling Track
Soltau, Hagen; Shafran, Izhak; Wang, Mingqiu; Rastogi, Abhinav; Han, Wei; Cao, Yuan
11:20 - 11:40 Best Paper
Adapting Text-based Dialogue State Tracker for Spoken Dialogues
Yoon, Jaeseok; Hwang, Seunghyun; Han, Ran; Bang, Jeong-Uk; Kim, Kee-Eung
Track 4 talks: Robust and Multilingual Automatic Evaluation Metrics for Open-Domain Dialogue Systems
Session Chair: Luis Fernando D'Haro
11:40 - 12:00 Track 4 Overview
Overview of Robust and Multilingual Automatic Evaluation Metrics for Open-Domain Dialogue Systems at DSTC 11 Track 4
Rodríguez-Cantelar, Mario; Zhang, Chen; Tang, Chengguang; Shi, Ke; Ghazarian, Sarik ; Sedoc, João; D'Haro, Luis Fernando; Rudnicky, Alexander I.
12:00 - 12:20 Best Paper
Simple LLM Prompting is State-of-the-Art for Robust and Multilingual Dialogue Evaluation
Mendonça, John; Pereira, Patrícia; Moniz, Helena; Carvalho , Joao Paulo; Lavie, Alon; Trancoso, Isabel M
Lunch
12:20 - 14:00 Lunch break (lunch served from 12:30 to 13:30)
Track 5 talks: Task-oriented Conversational Modeling with Subjective Knowledge
Session Chair: Seokhwan Kim
14:00 - 14:20 Track 5 Overview
Task-Oriented Conversational Modeling with Subjective Knowledge Track in DSTC11
Kim, Seokhwan; Gella, Spandana; Zhao, Chao; Jin, Di; Papangelis, Alexandros; Hedayatnia, Behnam; Liu, Yang; Hakkani-Tur, Dilek Z
14:20 - 14:40 Best Paper
Leveraging Ensemble Techniques and Metadata for Subjective Knowledge-grounded Conversational Systems
Joo, Seongho; Lee, Kang-il; Min, Kyungmin; Shin, Joongbo; Han, Janghoon; Won, Seungpil; Jung, Kyomin
Panel Discussion
14:40 - 15:30 Panel Discussion
Moderator: Yun-Nung (Vivian) Chen
Panelists: Luis Fernando D'Haro, Matthew Henderson, Chiori Hori, Alon Lavie, Weiyan Shi
15:30 - 16:00 Coffee break
Future DSTC & Closing
Session Chair: Michel Galley
16:00 - 16:30 Proposals for DSTC12
16:30 - 16:40 Closing Remarks
Poster Sessions
16:40 - 17:40 In-person Poster Session (A0 portrait size posters) – papers from all tracks
17:40 - 18:40 Virtual Poster Session (Discord server) – papers from all tracks
All posters (in-person and virtual):
2 - Exploring Prompt-based Multi-task Learning for Multimodal Dialog State Tracking and Immersive Multimodal Conversation
Chen, Yirong; Li, Ya; Wang, Tao; Xing, Xiaofen; Xu, Xiangmin; Liu, Quan; Liu, Cong; Hu, Guoping
4 - CopyT5: Copy Mechanism and Post-Trained T5 for Speech-Aware Dialogue State Tracking System
Park, Cheonyoung; Ha, Eunji; Jeong, Yewon; Kim, Chi-young; Yu, Haeun; Sung, Joo-won
5 - Multi-Stage Coarse-to-Fine Contrastive Learning for Conversation Intent Induction
Chu, Caiyuan; Li, Ya; Liu, Yifan; Gu, Jia-Chen; Liu, Quan; Ge, Yongxin; Hu, Guoping
6 - OLISIA: a Cascade System for Spoken Dialogue State Tracking
Jacqmin, Léo; Druart, Lucas; Estève, Yannick; Favre, Benoît; Rojas, Lina M; Vielzeuf, Valentin
8 - DORIC : Domain Robust Fine-Tuning for Open Intent Clustering through Dependency Parsing
Lee, Jihyun; Seo, Seungyeon; Kim, Yunsu; Lee , Gary Geunbae
9 - Multi-Task Learning for Ambiguous Candidate Identification with Pre-trained Model
Jang, Daesik; Choi, Hyewon
10 - Speech-Aware Multi-Domain Dialogue State Generation with ASR Error Correction Modules
Jiang, Ridong; Shi, Wei; Wang, Bin; Zhang, Chen; Zhang, Yan; Pan, Chunlei; Kim, Jung Jae; Li, Haizhou
13 - Contrastively Pretrained Vision-Language Transformers and Domain Adaptation Methods for Multimodal TOD Systems
Chang, Youngjae; Kim, Doo Young; Kim, Jinyoung; Kim, Keunha; Cha, Hyunmook; Min, Suyoung; Ko, Youngjoong; Lee, Kye-Hwan; Park, Joonwoo
15 - Analysis of Utterance Embeddings and Clustering Methods Related to Intent Induction for Task-Oriented Dialogue
Park, Jeiyoon; Jang, Yoonna; Lee, Chanhee; Lim, Heuiseok
17 - Multi-View Zero-Shot Open Intent Induction from Dialogues: Multi Domain Batch and Proxy Gradient Transfer
Koh, Hyukhun; Pyun, Haesung; Yang, Nakyeong; Jung, Kyomin
21 - Towards Optimizing Pre-trained Language Model Ensemble Learning for Task-oriented Dialogue System
Zhu, Zhiyuan; Liao, Yusheng ; Chen, Zhe; Wang, Yu; Guan, Yunfeng
23 - Enhancing Task-Oriented Dialog System with Subjective Knowledge: A Large Language Model-based Data Augmentation Framework
Jung, Haein; yeen, heuiyeen; Lee, Jeehyun; Kim, Minju; Bang, Namo; Koo, Myoung-Wan
24 - Parallel Corpora Alignment Framework for Multilingual and Robust Automatic Dialogue Evaluation
Wang, Xinglin; Shi, Jiayi; Yuan, Peiwen; Li, Kan
25 - Semantic data augmentation for meaning maintenance on Task-Oriented Conversation with Large-size Language Model
Lee, Jaehwan; Son, Kwanyoung; 유, 은경; Kim, Eugene
26 - Ensemble Method via Ranking Model for Conversational Modeling with Subjective Knowledge
Huang, Xin; Tan, Kye Min; Duan, Richeng; Zou, Bowei
28 - Exploring Back Translation with Typo Noise for Enhanced Inquiry Understanding in Task-Oriented Dialogue
Lee, Jihyun; Kim, Junseok; Lee , Gary Geunbae
29 - Leveraging Few-Shot Data Augmentation and Waterfall Prompting for Response Generation
Krause, Lea; Báez Santamaría, Selene; van der Meer, Michiel; Khurana, Urja
30 - Three Ways of Using Large Language Models to Evaluate Chat
Plátek, Ondřej; Dusek, Ondrej; Schmidtova, Patricia; Hudecek, Vojtech; Lango, Matheusz
32 - A Difference-aware Ensemble Method for Task-oriented Dialogue with Subjective Knowledge
Ke, Changxin; Sun, Churui; Ma, Longxuan; Zhang, Wei-Nan; Liu, Ting