{"id":"self-critique-harmful","text":"LLM revision based on self-critique makes answers worse: Sonnet -11pp, Flash -21pp, Pro -56.5pp. Self-critique fails because the same model that made the error evaluates the error","truth_value":"IN","source":"repo:beliefs-pi/entries/2026/05/06/generate-and-critique-llms-are-half-a-mind.md","source_url":"","source_hash":"","justifications":[],"dependents":["generate-and-critique"],"metadata":{},"explanation":{"steps":[{"node":"self-critique-harmful","truth_value":"IN","reason":"premise"}]}}