Hacker News new | ask | show | jobs
by rgoulter 415 days ago
Since LLMs are sometimes wonderfully useful, and sometimes not, I'd suggest effective use involves figuring out in which cases it's likely to succeed, which it's likely to fail.

For example, the mentioned graph has "initial prompt with iterative tweaks", followed by iterations of 'starting from scratch'. -- I don't understand why you'd think "this is an ineffective way of doing things", and then keep doing it.

Describing LLMs as "slot machines" seems like the author has no curiosity about the shape of what LLMs can/can't do.

2 comments

Answer is useful as a suggestion, and doesn't need to be factually correct: Good

Answer is useful as is, and needS to be factually correct: Bad

I do think its a bit slot machine like because the output isnt deterministic.
and they prompt you to add more credits, it's seamless to spend more - slot machine is referring to the experience using it, not what is happening technically