SCAI 2026

SIGIR 2026, July 24, Melbourne, Australia

The Search-Oriented Conversational Artificial Intelligence workshop (SCAI) is 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. SCAI is an established venue with a long-standing tradition of building bridges and integrating expertise from diverse research communities.

SCAI celebrates its 10th anniversary this year, and we warmly invite our core research community to join us!

Since our first workshop started back at ICTIR 2017 in Amsterdam, we have come a long way. We would like to use this milestone to reflect on that journey together. With the advent of large language models, conversational AI has become a dominant paradigm for search-intensive tasks. However, despite the vast success of conversational AI, there are major shortcomings in existing solutions that offer promising opportunities for the next breakthroughs which we would like to promote further. The focus of this year’s edition is on personalization of conversational search systems, and features a dedicated session for the former TREC shared task “Interactive Knowledge Assistance Track” (iKAT) reintroduced this year at SCAI.

Alongside a panel discussion, invited presentations, and keynote talks from leading industry representatives, the workshop will include a lively poster session and a dedicated break-out session with hands-on evaluation of the top-notch conversational AI systems. We are planning a full-day workshop that is dense, highly engaging, and community-driven.

Important Dates

  • Submission Deadline: April 27, 2026 (AoE)
  • Author Notification: May 21, 2026
  • Workshop Date: July 24, 2026

Program

We aim for a highly interactive program. Details will be released soon.

Call for Papers

We invite submissions of original and already published work.

Relevant topics

We invite submissions of original as well as recently published work related to the following topics:

  • Personalisation: Methods for user modeling, preference learning, and context-aware conversational search systems. Includes privacy, transparency, fairness, and evaluation benchmarks for personalized systems.

  • Multi-hop reasoning and complex information needs: Retrieval and reasoning over collections of documents and heterogeneous sources. Includes approaches for planning, provenance-based answering, and long-horizon state tracking.

  • Integrating tools: Tool-augmented conversational search, including integration of APIs, databases, web search, and agentic components. Includes orchestration and routing strategies, as well as end-to-end evaluation of robust and reliable tool use.

  • Learning from feedback: Learning from implicit and explicit feedback to improve conversational search quality. Includes human-in-the-loop training and evaluation, LLM-as-a-judge approaches for scalable assessment, and alignment trade-offs.

  • Multimodality: Conversational search over (un)structured text, images, audio, and video. Includes multimodal retrieval-augmented generation, systems designed for interaction across modalities, and evaluation through benchmarks and user studies.

  • Open/closed, small/large LMs: Comparative studies of model families and scales for conversational search, with a focus on efficiency and deployment constraints. Includes reproducibility, transparency, and safety considerations across settings.

Submission

We invite submissions of long papers (8 pages), short papers (4 pages), as well as paper proposals (2 pages) describing ongoing research. Submissions can be original papers or recently accepted papers.

Submissions should be made via EasyChair. Authors should use the official ACM proceedings templates (LaTeX, Word, and Overleaf). Please use the sigconf proceedings format, available from the ACM website. Submission to this workshop implies agreement with, and compliance to, the ACM and SIGIR 2026 Policies and Requirements.

iKAT Shared Task

We invite participation in our shared task about personalized conversational information access called “Interactive Knowledge Assistance Track (iKAT)”, formerly hosted at TREC. All details about the shared task can be found on the iKAT website.

Organisers

  • Philipp Christmann, CISPA Helmholtz Center for Information Security
  • Roxana Petcu, University of Amsterdam
  • Sneha Singhania, Max Planck Institute for Informatics
  • Mohammad Aliannejadi, University of Amsterdam
  • Marcel Gohsen, Bauhaus-Universität Weimar
  • Svitlana Vakulenko, Vienna University of Economics and Business

The organizers acknowledge that this workshop was conceived on the lands of the First Nations peoples of the Kulin Nation, Kanien’kehá:ka (Mohawk) Nation, and the Pennacook, Abenaki, and Wabanaki Peoples. The workshop will be co-located with ACM SIGIR 2026 and hosted on the unceded lands of the Woi wurrung and Boon wurrung language groups of the Eastern Kulin Nation. We acknowledge their Ancestors and Elders, past, and present, and we respectfully acknowledge their connection to Country.