CARES: A Conversational AI System for Regulation-Grounded Safety Reporting in Construction Education
3.00T1 sourcearXiv cs.HC
Source record
Published by arXiv cs.HC (T1 source). The original is at https://arxiv.org/abs/2609.19429.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryCARES is a conversational AI system for construction safety reporting that combines proactive multi-agent dialogue, retrieval-augmented generation with hybrid retrieval, and automated structured report generation. A 15-student evaluation reported 0.74 faithfulness and 1.00 answer relevance, with initial retrieval ranking noted as a weakness.
Why it mattersDomain-specific but documented: shows how multi-agent dialogue plus RAG can be wired into a regulated reporting workflow, with measured numbers and a clear limitation on retrieval ranking.
Cited by
No citations on record.
