Rumours say you do something like:

  Download every github repo
    -> Classify if it could be used as an env, and what types
      -> Issues and PRs are great for coding rl envs
      -> If the software has a UI, awesome, UI env
      -> If the software is a game, awesome, game env
      -> If the software has xyz, awesome, ...
    -> Do more detailed run checks, 
      -> Can it build
      -> Is it complex and/or distinct enough
      -> Can you verify if it reached some generated goal
      -> Can generated goals even be achieved
      -> Maybe some human review - maybe not
    -> Generate goals
      -> For a coding env you can imagine you may have a LLM introduce a new bug and can see that test cases now fail. Goal for model is now to fix it
    ... Do the rest of the normal RL env stuff

▲ NitpickLawyer 5 hours ago | parent [-]

The real real fun begins when you consider that with every new generation of models + harnesses they become better at this. Where better can mean better at sorting good / bad repos, better at coming up with good scenarios, better at following instructions, better at navigating the repos, better at solving the actual bugs, better at proposing bugs, etc.

So then the next next version is even better, because it got more data / better data. And it becomes better...

This is mainly why we're seeing so many improvements, so fast (month to month, from every 3 months ~6 monts ago, from every 6 months ~1 year ago). It becomes a literal "throw money at the problem" type of improvement.

For anything that's "verifiable" this is going to continue. For anything that is not, things can also improve with concepts like "llm as a judge" and "council of llms". Slower, but it can still improve.

▲

alex43578 5 hours ago | parent | next [-]

Judgement-based problems are still tough - LLM as a judge might just bake those earlier model’s biases even deeper. Imagine if ChatGPT judged photos: anything yellow would win.

	▲	NitpickLawyer 4 hours ago \| parent \| next [-]
		Agreed. Still tough, but my point was that we're starting to see that combining methods works. The models are now good enough to create rubrics for judgement stuff. Once you have rubrics you have better judgements. The models are also better at taking pages / chapters from books and "judging" based on those (think logic books, etc). The key is that capabilities become additive, and once you unlock something, you can chain that with other stuff that was tried before. That's why test time + longer context -> IMO improvements on stuff like theorem proving. You get to explore more, combine ideas and verify at the end. Something that was very hard before (i.e. very sparse rewards) becomes tractable.
	▲	cindyllm 5 hours ago \| parent \| prev [-]
		[dead]

▲

losvedir 4 hours ago | parent | prev [-]

Yeah, it's very interesting. Sort of like how you need microchips to design microchips these days.