EY re-envisions RAG around multimodal knowledge graphs to improve accuracy
Most RAG systems miss 40% of enterprise data. EY's multimodal knowledge graph approach changes that.
Why it matters
EY research identifies a critical gap in how enterprises deploy RAG—conventional text-only systems leave charts, tables, and visual data untapped. This academic/research finding has immediate relevance for CTOs and AI leaders evaluating RAG architectures for production.
The key facts
9 to knowEY research on RAG limitations
Conventional RAG systems retrieve only text
Enterprise documents contain critical facts in charts and tables
Multimodal knowledge graphs proposed as solution
Focus on improving accuracy of RAG implementations
Conventional RAG systems built primarily for text retrieval
Enterprise documents contain critical facts in charts, tables, and visual formats
EY proposes multimodal knowledge graphs as improved RAG architecture
Focus on improving accuracy and data utilization in enterprise AI deployments
Go to the source
SiliconAnglesiliconangle.com
Publisher excerpt: Retrieval-augmented generation is a standard way to ground large language models in enterprise information, but new research from EY, the business name of Ernst & Young LLP, says most implementations overlook a lot of useful data. Conventional RAG systems are built mainly to retrieve text.…