4 ms·
> All models exhibited limitations in spatial planning when constructing multi-section factories. Common failures included placing entities too close together,
by owenpalmer 2y ago
> All models exhibited limitations in spatial planning when constructing multi-section factories. Common failures included placing entities too close together, not allocating space for connections, or incorrect inserter placement
It makes sense why LLMs are bad with spatial reasoning. Not a lot of training data for it. I wonder what additional reasoning abilities will emerge when spatial reasoning is solved.
- wordpad 2y agoHow is there not a lot of special data? Isnt it literally infinite via even the simplest simulator? You could generate an unlimited training set just by implementing tik tac toe on an unbound grid, for example, in like 10 lines of code.
- owenpalmer 2y agoSynthetic data will play I big role, yes. There's other challenges though, like how verbal descriptions of objects would affect their spatial behavior. Building a generalized simulator that combines those modalities is hard. In this particular case with Factorio, I suspect generating the synthetic data would be easier, since the rules of the environment are relatively simple and well defined, with quantifiable outcomes.