[CFP] [Extended Deadline] IJCAI-ECAI 2026 Workshop: AI Modeling for Disappearing Knowledge (AIM4DK)
EmpAI
AI Modeling for Disappearing Knowledge (AIM4DK)
in conjunction with IJCAI-ECAI 2026 (August 15-21, Bremen, Germany)
New Submission Deadline: May 17, 2026 (23:59 AoE)
Overview
Modern AI systems are increasingly built on knowledge that is dynamic, unstable, and ephemeral. Web content disappears, scientific facts develop, datasets become outdated, and model training corpora cannot always be retrieved or verified. This raises a fundamental challenge: How can AI systems remain reliable when their underlying knowledge is incomplete, drifting, or no longer accessible?
This workshop explores AI under temporal uncertainty, focusing on the intersection of: Temporal drift (distribution and knowledge shift over time), Knowledge loss (missing or unverifiable sources), and Data ephemerality (disappearing or changing data). We aim to move beyond static assumptions of knowledge and explore robust, accountable, and temporally-aware AI systems, addressing both technical and epistemological questions of trust, verification, and reproducibility.
Topics of Interest:
We invite submissions including (but not limited to):
Data fragility and preservation
Dataset decay, disappearing web sources, and link rot in public resources
Data versioning, web archiving, and long-term dataset preservation
Symbolic and neuro-symbolic approaches to tracking knowledge provenance
Model robustness under temporal drift
Temporal drift and concept shift in machine learning models
Time-aware evaluation methods and protocols for AI reliability
Reasoning under uncertainty caused by missing or outdated context
Foundations, logic, and societal impact
Theoretical frameworks for learning under time-sensitive or disappearing data
Societal risks of "ghost knowledge" in law, science, and public policy
Cross-disciplinary studies of temporal degradation in real-world applications
LLMs and Dynamic Knowledge
Knowledge representation and updating in large language models
Hallucination and reliability under outdated or unverifiable information
Evaluation of LLMs under temporal drift and knowledge decay
Important Dates:
Paper submission deadline: May 17, 2026 (23:59 AOE)
Notification to authors: June 5, 2026 (23:59 AOE)
Camera-ready deadline: July 1, 2026 (23:59 AOE)
Workshop date: Aug 15-17, 2026 (exact date TBA)
Submission Guidelines:
We welcome submissions of full papers, short papers, work-in-progress papers, and vision or position papers relevant to the workshop topics.
The workshop uses OpenReview as the review platform
Submissions must be prepared using the standard IJCAI-ECAI 2026 Author Kit and submitted as a single PDF file
Long papers: up to 7 pages (main text)
Short papers: 2-4 pages (main text)
References and appendices do not count toward the page limit
Reviewing will be double-blind, and submissions must be anonymized accordingly
We also welcome submissions of previously published or accepted work relevant to the workshop topics. Such submissions should be in the form of an extended abstract or adapted version, clearly indicating the original publication.
Publication
This workshop is non-archival. Accepted papers will be made publicly available on OpenReview and hosted on the workshop website. Authors may opt out of public release. Submissions to this workshop can be further submitted to archival venues.
Attendance
AIM4DK is a fully in-person workshop. At least one author of each accepted paper is expected to attend the workshop and present the work. Accepted papers will be presented as posters and/or contributed talks.
Large Language Model (LLM) Policy
In line with the IJCAI-ECAI 2026 policies for the main track, AIM4DK does not allow AI systems to serve as primary authors or reviewers of any submission. Human authors are fully responsible for the content and originality of their submissions.
Contacts:
For further questions or concerns, please contact the organizing committee at aim4...@gmail.com