> to OpenAI's model gpt-4o-mini

Why a model specifically distilled down for logical reasoning tasks? I would expect larger models to produce a wider variety of outputs.

▲

TeMPOraL 16 hours ago | parent [-]

That, plus the quoted text basically says the model homed in on the monomyth (Hero's journey) structure; while the pattern was identified and named by a 20th century American writer, the pattern itself is common and as ancient as it gets. Wouldn't really call it anglo-american bias.

The monomyth is also writing 101 these days, and considered the default structure you can and should use if you have little experience writing stories, so naturally it'll be a high-probability result of an LLM prompted to write a story - especially prompted in a way that implies the user is inexperienced at writing and needs a result suitable for an inexperienced writer.

	▲	flir 15 hours ago \| parent [-]
		> a protagonist lives in or returns home to a small town and resolves a minor conflict by reconnecting with tradition and organising community events That's... not the Hero's Journey? (The same study run against Claude Opus would be interesting - if we're going to test models, we might as well play to their strengths. My prediction: better writing, not better plotting).