Welcome to CONVERSATIONS 2023, a two-day workshop on chatbot research, applications, and design. This is the 7th CONVERSATIONS workshop. With the surge of interest in chatbots powered by large language models over the last few months, the workshop is possibly more relevant than ever before. Join in for sharing and collaborating between researchers and practitioners in the field.
The workshop is located in Oslo, Norway. However, to accommodate for participation also for those not able to be there in person, the workshop will include some online sessions as part of the full on-site program. We look forward to both onsite and virtual contributions.
- Date: November 22-23
- Location: University of Oslo, Norway
- Workshop format: On-site and online
- Paper submission deadline: September 10
Researchers and practitioners are encouraged to submit papers and proposals for inclusion in the workshop, considering the following submission categories:
- Full papers: Empirical studies, theoretical contributions, design research, or presentations of chatbot systems and applications that advance the state-of-the-art (12-16 pages, Springer LNCS format). To be presented at the workshop and published in the Springer LNCS workshop proceedings.
- Position papers and project presentations: Presentations of exciting research projects, sharing of preliminary results and work in progress, or author positions on open issues (3-6 pages, Springer LNCS format). Research project presentations can be entirely focused on chatbot research or address research including major chatbot components. To be presented at the workshop and published at the workshop webpage.
- Groupwork proposals: Proposals for 2-hour groupwork sessions at the workshop. May for example concern design exercises, discussions of current topics, collaboration on theoretical or applied problems, challenges, or discussions of future research project ideas. Arranging a groupwork is an excellent opportunity to draw attention to issues of importance to the chatbot research community. For submission, a brief description of the groupwork proposal is needed (1-3 pages Springer LNCS format). Proposers of accepted groupworks are provided space to run these at the workshop and the groupwork proposal is published at the workshop webpage.
- September 10: Submission deadline
- October 1: Author notification
- October 24: Submission of revised papers
- November 22-23: Workshop
KEY TOPICS OF INTEREST
Building on the results from previous CONVERSATIONS workshops, the following topics are of particular interest to the workshop. Contributions concerning applications of large language models such as the GPT family, are warmly welcome, as are contributions on applications combining information retrieval approaches and large language model approaches.
1. Chatbot users and implications. How to understand chatbot and large language model user groups and their context of use? How may the uptake and use of chatbots and large language models impact individual users, groups of users, organizations and society at large?
2. Chatbot user experience, design, and evaluation. How to understand and improve the user experience of chatbots and large language model applications? How to design conversational interactions conductive to novel and improved user experiences? How multimodality impacts the user experience of chatbots and large language model applications? How to conduct user-centred evaluations and measurement?
3. Chatbot frameworks and platforms. Knowledge and experiences on current and future frameworks and platforms for chatbot development and delivery. For example, pertaining to data access, NLP, and context awareness. Chatbot training. Chatbots that learn through use. Multimodal chatbot interaction. Chatbots leveraging large language models.
4. Chatbots for collaboration. How to understand and design chatbots and large language model applications for collaboration in groups or teams of users? How to design for networks of humans and intelligent agents? Relevant collaboration domains may include social networks, teamwork, or service provision.
5. Democratizing chatbots – chatbots for all. How to use chatbots and large language models to improve availability, accessibility, effectiveness, and efficiency to information and services? How can chatbots bridge digital divides. How to design inclusive chatbots with and for diverse user groups?
6. Ethics and safety implications of chatbots and large language models. Design for ethics, safety, and trust in chatbots and large language model applications. Considerations on privacy, bias and discrimination. Mitigation of unethical use and analysis of potential harms (e.g., unintended offensive speech, misinformation spreading). Ethical aspects of the development of large language models used in chatbots.
7. Leveraging advances in AI technology and large language models. Recent progress on AI technology underpinning chatbots, including advances in large open-domain language models, opens new opportunities for chatbot use, including general purpose discussions, information search, question answering and the like. We particularly welcome papers addressing chatbot applications leveraging advances in AI technology and large language models, , e.g., for creative work, education and coding support, as well as novel aspects and challenges concerning their use such as prompt engineering or the issue of ‘hallucinations’ in large language models.
The set of relevant application areas for chatbots is increasing. We hope at the workshop to include work in areas such as the following:
- Government and public service. Providing information and services to citizens or supporting government service provision.
- At the workplace. The use of chatbots and large language models in decision support tools, as interface for business intelligence applications, and to support general office work.
- In the home. The use of chatbots and large language model applications to search for and access information, services, and home applications. Applications for home assistants.
- Customer service. Chatbots and large language model applications used by service providers to engage with users – for support and information purposes.
- Education. How chatbots and large language models can support teaching or training, to assist with administrative tasks, or to bring students together. Challenges with educational applications and how to mitigate these.
- Software development. Chatbots and large language model applications for coding support. Opportunities such as efficiency improvements, benefits for non-experts. Challenges such as security and code quality.
- Health and therapy. Chatbots and large language model applications for health advice, medical counselling, therapeutic programs, or physical training programs. Opportunities and challenges.
- Media and journalism. Chatbots and large language model applications supporting news reporting and consumption. Use cases and critical discussions.
- Creative industries. Chatbots and large language model applications to engage consumers in creative industries services or events.
- Ecommerce and marketing. Chatbots and large language models supporting marketing for brands or brokering and sales of goods and services.
The workshop organising committee are:
- Asbjørn Følstad, SINTEF, Norway
- Theo Araujo, University of Amsterdam, the Netherlands
- Symeon Papadopoulos, Centre for Research and Technology Hellas, Greece
- Effie L.-C. Law, University of Leicester, UK
- Ewa Luger, University of Edinburgh, UK
- Sebastian Hobert, University of Göttingen, Germany
- Morten Goodwin, University of Agder, Norway
- Petter Bae Brandtzaeg, University of Oslo and SINTEF, Norway