Remix clone Hacker News

new | show | ask | jobs Github

	▲	noosphr 7 hours ago
		`People defer thinking about what correct and incorrect actually looks like for a whole wide scope of scenarios and instead choose to discover through trial and error.` LLMs are _still_ terrible at deriving even the simplest of logical entailment. I've had the latest and greatest Claude and GPT derive 'B instead of '(not B) from '(and A (not B)) when 'A and 'B are anything but the simplest of English sentences. I shudder to think what they decide the correct interpretations of a spec written in prose is.
	▲	layer8 24 minutes ago \| parent \| next [-]
		Lisp quotes are confusing in prose.
	▲	Kiro 5 hours ago \| parent \| prev \| next [-]
		I would love to see a prompt where it fails such a thing. Do you have an example?
	▲	0x696C6961 6 hours ago \| parent \| prev [-]
		Still better than my coworkers ...