4 ms·
What do you think a dog or cat does that can't be replicated by LLMs?
by snapcaster 2y ago
What do you think a dog or cat does that can't be replicated by LLMs?
- jfengel 2y agoI don't know. I know that any particular thing I say could be met with an LLM that did that thing. But at the moment an LLM doesn't seem to be capable of coding itself up that way. Yet.
- snapcaster 2y agoI guess that was kind of my question when I thought about what you said. I could think of any things that a cat or dog do that _current_ LLMs can't do but nothing that seems fundamentally out of reach
- skydhash 2y agoAct on its own?
- snapcaster 2y agoWhat does it mean for something to "act on its own"? Every action taken is in some part due to what's happening in the environment right? I would argue that nothing "acts on it's own"
- charlescurt123 2y agocontinually learn with a fixed computational budget over massively extended task horizons.
- arminiusreturns 2y agoNo, but it could help me make such a thing.
- charlescurt123 2y agoPossibly, but as of now it's a completely unsolved problem and to my knowledge nobody has shown even a tiny model being able to perform it. Based on the top page today I may even be able to make the argument we can't even simulate the abilities of a fruit fly. The absolute frontier models can perform only a fraction of a fraction of what a typical work day looks like for a human. I calculated the chance that a frontier model today has about a 1^-29 chance of performing a single day of connected tasks based on GAIA benchmarks.