University of South Carolina develops AI system for academic policy assistance

Researchers have developed Carolina Guide, a retrieval-augmented generation (RAG) system designed to assist the University of South Carolina (USC) with academic policy inquiries. This multi-agent system incorporates institutional guardrails to ensure answers are grounded in policy and supported by citations, while also refusing unsafe requests like personalized advising. Evaluations demonstrated high retrieval success rates and effective refusal of adversarial queries, highlighting the need for specialized RAG architectures that prioritize safety and transparency in high-stakes institutional contexts. AI

IMPACT This system demonstrates a specialized application of RAG for institutional policy guidance, potentially improving student access to information and reducing administrative load.

RANK_REASON The cluster describes a research paper detailing a new AI system for academic policy assistance. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• arXivLabs
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• ScienceCast
• University of South Carolina
• University of Southern California

• product
• safety
• paper

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →