This workshop, centered on “Medical AI and Causal Explanations,” builds upon the achievements of NTCIR-18 HIDDEN-RAD (Hidden Causality Inclusion in Radiology Report Generation) task to share causal reasoning and explanation-generation techniques across multiple domains.
Objectives:
Expand causal-explanation research—originating from medical radiology reports—into law, science, finance, and other fields.
Share advancements in AI explainability, hallucination detection, and correction.
Key Themes:
Causal reasoning frameworks for professional domains (e.g., radiology, legal texts, financial reports).
Explanation generation techniques (Chain-of-Thought, RAG, hybrid knowledge integration).
Hallucination (unfounded or incorrect explanations) detection and mitigation.
Cross-domain transfer of causal-explanation methods.