Modern vision-language models allow documents to be transformed into structured, computable representations rather than lossy text blobs.

Traditional storytelling focuses on emotion, arc and persuasion. That still matters for humans. But LLMs also require ...