---
format: "aidr-story-markdown/v1"
id: "22c2e279792d4503b3ac8b479d16295f366290ac0d201c308faa52a24c5bd78d"
canonical_url: "https://aidr.today/22c2e279?lang=en"
title: "Models ranked by normative score across twelve paradigms, once neutral and once human-primed"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-04T09:21:18.000Z"
category: "Research"
topics: ["llm"]
source_urls: ["https://cognit.rajtilak.tech/","https://lobste.rs/s/uuxij7/models_ranked_by_normative_score_across"]
summary: "How carefully language models answer. Twelve puzzles, scored by another model."
---

# Models ranked by normative score across twelve paradigms, once neutral and once human\-primed

> [Open the canonical story](<https://aidr.today/22c2e279?lang=en>)

**Published:** 2026-10-04T09:21:18.000Z
**Category:** Research
**Topics:** llm

## Summary

How carefully language models answer\. Twelve puzzles, scored by another model\.

## Sources

- [Story source](<https://cognit.rajtilak.tech/>)
- [Discussion](<https://lobste.rs/s/uuxij7/models_ranked_by_normative_score_across>)