Hop-Decayed Influence: New Vulnerabilities of Structural Auxiliary Indexing in GraphRAG Pipelines with LLMResearchOct 5, 2026
EviDent-CBCT: Evidence-Bottlenecked Report Generation from Dental CBCT under Non-Exhaustive Report SupervisionResearchOct 5, 2026
Fast Models, Slow Evidence: A Paired and Self-Audited Evaluation of System-1 Decision Models for LLM Agent HarnessesResearchOct 5, 2026
The AI Risk Observatory: What Can We Learn from AI Disclosures in Annual Reports About Societal Resilience?ResearchOct 5, 2026