SCAI 2021

SCAI 2021

Online, 8 October

SCAI 2021 is intended as a discussion platform on Conversational AI for intelligent information access, bringing together researchers and practitioners across natural language processing, information retrieval, machine learning and human-computer interaction fields. Among other topics, we will discuss design, evaluation and human factors in relation to automating information-seeking dialogues. The workshop also featured a shared task on Conversational Question Answering.

SCAI is already an established venue with a long-standing tradition of building bridges and integrating expertise from diverse research communities. It was previously organised in the context of ICTIR 2017, EMNLP 2018, TheWebConf 2019, IJCAI 2019 and EMNLP 2020. To access the materials of the previous five editions of SCAI navigate to https://scai-workshop.github.io

While 2021 marks a 6th iteration of SCAI, this year, the workshop now focuses on the discussion between researchers from different fields. It is therefore organized as a strictly non-archival venue, as an opportunity to present papers accepted to other venues in an interdisciplinary meeting specifically focused on search-based conversational AI.

The workshop is organized as an online event, with free participation.

SCAI 2021 Program

Check out the recordings of SCAI’21 on our YouTube channel

Opening (09:40-09:50 CET)

Conversational Search Evaluation Session (10:00-10:50 CET)

• Zeyang Liu (University of Nottingham) Meta-evaluation of Conversational Search Evaluation Metrics TOIS 21

• Aldo Lipani (University College London) How Am I Doing?: Evaluating Conversational Search Systems Offline TOIS 21

• Alexandre Salle (Federal University of Rio Grande do Sul) Studying the Effectiveness of Conversational Search Refinement through User Simulation ECIR 21

• Hsien-Chin Lin (Heinrich-Heine-Universitat Dusseldorf) Domain-independent User Simulation with Transformers for Task-oriented Dialogue Systems SigDial 21

User Satisfaction & Dialogue Breakdown Session (11:00-11:50 CET)

• Shuo Zhang (Bloomberg) Simulating User Satisfaction for the Evaluation of Task-oriented Dialogue Systems SIGIR 21

• Leon-Paul Schaub (Universite Paris-Saclay) Defining And Detecting Inconsistent System Behavior in Task-oriented Dialogues TALN 21

• Katsuhide Fujita (Tokyo University of Agriculture and Technology) Dialogue Act-based Breakdown Detection in Negotiation Dialogues EACL 21

Poster Session + Lunch Break (12:00-14:00 CET)

Dialogue Personalization Session (14:00-14:50 CET)

• Sergey Volokhin (Emory University) You Sound Like Someone Who Watches Drama Movies: Towards Predicting Movie Preferences from Conversational Interactions NAACL 21

• Andrew Yates (University of Amsterdam) You Get What You Chat: Using Conversations to Personalize Search-Based Recommendations ECIR 21

• Marco Polignano (University of Bari) MyrrorBot: A Digital Assistant Based on Holistic User Models for Personalized Access to Online Services TOIS 21

• Lucie Flek (University of Marburg) Towards User-Centric Text-to-Text Generation: A Survey TSD 21

Question Answering Evaluation Session (15:00-15:50 CET)

• Abhilasha Ravichander (Carnegie Mellon University) NoiseQA: Challenge Set Evaluation for User-Centric Question Answering EACL 21

• Linqing Liu (University College London) Challenges in Generalization in Open Domain Question Answering

• Jifan Chen (University of Texas at Austin) Can NLI Models Verify QA Systems’ Predictions? EMNLP 21

• Vaibhav Adlakha (McGill University) TopiOCQA: Open-domain Conversational Question Answering with Topic Switching

SCAI-QReCC Shared Task on Conversational QA (16:00-16:50 CET)

Organisers:

• Leigh Clark, Swansea University
• Ondrej Dusek, Charles University
• Svitlana Vakulenko, University of Amsterdam

Shared Task Organisers:

• Maik Fröbe, Martin Luther University of Halle-Wittenberg
• Johannes Kiesel, Bauhaus-Universität Weimar
• Svitlana Vakulenko, University of Amsterdam

Steering Committee:

• Mikhail Burtsev, MIPT & DeepPavlov
• Heloisa Candello, IBM Research
• Aleksandr Chuklin, Google Research
• Milica Gasic, Heinrich-Heine-Universität Düsseldorf
• Evangelos Kanoulas, University of Amsterdam
• Julia Kiseleva, Microsoft Research
• Ahmet Baki Kocaballi, University of Technology Sydney
• Minha Lee, Eindhoven University of Technology
• Martin Porcheron, Swansea University
• Verena Rieser, Heriot-Watt University & ALANA AI
• Stefan Ultes, Mercedes-Benz Research
• Marilyn Walker, University of California Santa Cruz

Sponsors: