Discussion

yare

vessenes: Cool!From the prompt it looks like you don’t give the llms a harness to step through games or simulate - is that correct? If so I’d suggest it’s not a level playing field vs human written bots - if the humans are allowed to watch some games that is.

levmiseri: That’s true, I’m trying to figure out a better testing environment with a feedback loop.I did try letting the models iterate on the bot code based on a summary of an end-of-game ‘report’, but that showed only marginal improvements vs. zero-shot

DeathArrow: LLMs need to have feedback of the outcomes. Just like a human does.

vessenes: In my mind, I’d give it the following:Step(n) - up to n steps forwardRunTil(movement|death|??) - iterate until something happensBoard(n) - board at end of step nBoardAscii(n) - ascii rep of sameLog(m,n) - log of what happened between step m and nProbably all this could be accomplished with a state structure and a rendering helper.Do you let humans review opposing team’s code?

neondude: You should check out codingame.com It has similar battle based objectives