Workshop overview

Much of the knowledge organizations rely on is implicit: it lives in the routines, judgement, and experience of individual experts, and it is rarely written down. As workforces turn over and experts retire, this knowledge is lost before it ever reaches a document or database.

The information and knowledge management community has strong methods for extracting and organizing knowledge that already exists in text and data. Comparatively little attention is paid to the step before that: surfacing implicit knowledge from people and work practice so that it can be represented, integrated, and preserved at all.

KEEPER focuses on this gap. The workshop brings together research on an implicit-knowledge lifecycle with four stages:

Elicitation (capture and identify). Interactive and AI-supported methods to surface implicit knowledge from experts, including LLM-driven interviewing, dialogue agents, and conversational elicitation, and methods to identify which knowledge is implicit, valuable, and at risk.
Externalization (represent). Turning elicited knowledge into structured, reusable representations such as knowledge graphs, ontologies, and documented procedures, with attention to provenance and to what is lost in formalization.
Integration. Linking newly externalized knowledge with existing knowledge management systems, retrieval pipelines, and organizational repositories without creating silos or contradictions.
Preservation and organization. Maintaining elicited knowledge as durable organizational memory: versioning, curation, retrieval, and guarding against knowledge loss over time.

Throughout, KEEPER treats AI as assistive and human-in-the-loop, with the human expert as the source and validator of knowledge. The workshop complements the CIKM main tracks: it addresses knowledge that is not yet recorded at all, rather than extraction from existing data.
Topics of interest

We invite contributions on, but not limited to:

LLM-supported and interactive elicitation of implicit and expert knowledge (interviewing, dialogue agents, conversational elicitation)
Identifying which knowledge is implicit, valuable, and at risk within an organization
Externalization into knowledge graphs, ontologies, and documented procedures; provenance and loss in formalization
Integration of newly externalized knowledge with existing knowledge management and retrieval systems
Preservation, curation, versioning, and retrieval of knowledge as organizational memory
Human-in-the-loop and trustworthy AI for knowledge capture and validation
Knowledge loss through workforce turnover; knowledge management in SMEs and the public sector
Evaluation methods, datasets, benchmarks, and field studies, including honest negative or partial results from real deployments