A financial services company uses a large language model for an internal tool that summarizes complex regulatory documents for its legal team. The team reports that while the summaries are generally well-written, the model sometimes hallucinates non-existent clauses or misinterprets critical legal terms, posing a significant risk. To address this primary concern, which approach to improving the model should the company prioritize?

Scaling for Creative AI

A development team is working on improving a large language model for different applications. Match each development goal with the primary dimension of scaling it represents.