Call for Papers
We invite submissions of original work as well as vision and position papers to the 1st Workshop on Conversational Search for Complex Information Needs.
For any information needed that is not listed below, please submit questions to the Team.
Important Dates
- Paper submission deadline: February 1, 2025
- Notification of acceptance: February 21, 2026
- Workshop date: April 2, 2026 (co-located with ECIR, the European Conference on Information Retrieval)
Submission Types
We invite two main types of contributions:
-
Technical Papers (4–8 pages)
Novel research contributions, including models, algorithms, user studies, system analyses, benchmarks, or empirical evaluations. -
Position / Vision Papers (up to 4 pages)
Forward-looking, thought-provoking papers that articulate challenges, research agendas, or conceptual frameworks for conversational search and complex information needs.
Page limits are excluding references (you can adjust this if needed). The workshop will be non-archival, so authors are free to submit extended versions of their work to other venues.
Submission Guidelines
- Submissions should be original and not under review at another archival venue.
- Papers should follow the ECIR formatting guidelines (same style as the main conference).
- All submissions will be peer reviewed by the workshop Program Committee for relevance, quality, and potential to stimulate discussion. - At least one author of each accepted paper must register for the workshop and present their work.
Submission site: https://easychair.org/conferences?conf=coscinecir2026
Topics of Interest
We welcome (but are not limited to) contributions on the following topics:
- Complex Information Needs
- Modeling, detecting, and supporting complex, exploratory, and multi-step information needs
- Reasoning, planning, and synthesis across multiple documents or tools
- Handling implicit logical operations (negation, conjunction, disjunction, set operations) in conversational search
- Approaches that go beyond model training data or parametric knowledge (e.g., retrieval-augmented, hybrid systems)
- Data & Benchmark Generation
- Benchmarks for complex conversational search
- Synthetic data generation and augmentation for multi-turn, intent-aware dialogs
- Methodologies for evaluating new datasets against existing benchmarks
- Orchestration & Agentic Systems
- Adaptive orchestration of tools, retrievers, and models in conversational pipelines
- Agent-based approaches for deciding when/how to retrieve, plan, or ask clarification questions
- Optimization and evaluation of dynamic, modular systems
- Personalization & Adaptivity
- Personalized conversational search and recommendation
- User modeling over long-term, multi-session interactions
- Adaptive response styles, tones, and explanations for users with diverse backgrounds
- Fair and responsible personalization under bias and stereotype constraints
- User Experience & Interaction Design
- Studies of how users interact with conversational vs. traditional search
- Overreliance, trust calibration, and verification behavior
- Interfaces that help users understand, steer, and correct conversational agents
- Mixed-initiative and collaborative conversational search
- Bias, Fairness, Interpretability & Safety
- Bias detection and mitigation in conversational search and RAG systems
- Fairness, inclusivity, and societal considerations in conversational AI
- Faithfulness, explanation, and attribution in multi-step reasoning and retrieval
- Evaluation
- Evaluation methodologies for long-form answers, reasoning traces, and multi-turn interaction
- User-centered evaluation of safety, trustworthiness, personalization, and satisfaction
- Automatic and human evaluation frameworks for complex conversational tasks
We encourage interdisciplinary work and case studies from domains such as education, healthcare, finance, public services, and other high-stakes settings.
Presentation Format & Workshop Program
To encourage interaction and discussion, the workshop will combine:
- Keynote talks by invited experts
- Panel discussion on challenges and future directions in conversational search
- Poster session for all accepted papers
- One-minute “lightning” pitches for poster presentations
Accepted papers will be presented as posters, preceded by short pitches to the full audience. We plan a half-day workshop (approximately 3.5 hours plus breaks).