4 ms·
This sounds promising. Smaller but custom trained/tuned models would be ideal - works for the task without the overhead
by Havoc 3y ago
This sounds promising. Smaller but custom trained/tuned models would be ideal - works for the task without the overhead
- logicchains 3y agoIn my experience it doesn't work like that. It's like if you take an idiot and spend a bunch of time training them, they'll still perform much worse than a moderately intelligent person with much less training. And smaller models can be pretty idiotic.
- sysstemlord 3y agoThen why do Chess AI perform much better than LLMs trying to play chess.
- Gasp0de 3y agoBecause they do completely different things? They literally have nothing to do with each other. Why do planes fly better than ships if ChatGPT can't do math?
- RugnirViking 3y agoProvided that the problem is suited to the strengths of an LLM at all. An example might be a small ai custom trained on documentation for libraries. You ask it a question like "how do I make the background move with parallax effect when you move the cursor". It's a little ambiguous, high-level concept, and probably not a single function. Small ai: likely makes up a function or suggests a single function which isn't sufficient. Refuses to budge from its answer or apologies and gets confused Large LLM: able to actually understand the question, combine several functions. If it doesn't work you can tell it why and it fixes it
- solumunus 3y agoWhy would a language model be good at playing chess?
- kohlerm 3y agoWhy no? Chess notation is text. But the problem is that LLMs are not that good for problems which require evaluation of a search tree. Also leading chess engines such as lc0 are without search better than 90+x% of All humans
- seanhunter 3y agoBecause there’s a world of difference between a reinforcement learning trained special purpose model and asking a general purpose large language model to have a go at something.
- PeterStuer 3y agoBecause they have an explicit model of chess and specific heuristics for learning chess. An LLM could have picked up some chess patterns through osmosis, but it can not reason explicitly in the domain.
- kohlerm 3y agoNo"they" (lc0) don't have specific heuristics
- johnlyd 3y agoAgreed.
- pzo 3y agoAgreed, mostly in real life we have specialized experts. You go to doctor to ask about health related stuff, you ask your colleague who is expert in ML only about ML and probably not much about mobile development. Instead if having ML that is expert at coding in all languages probably would be better to allow to switch context. Native mobile dev LLM? - e.g. train only on swift, objc, kotlin, java, c, c++ code Python dev LLM? - train only on python, c, c++, rust code Would this way final model be smaller, faster and maybe event better?
- treprinum 3y agoUnfortunately, for transformer-based LLMs the magic starts only when they are trained by more that 10^22 TFlops (preferably 10^24) so smaller models might not cut it even for fine-tuned tasks.
- visarga 3y agoAt medium size (13B) Microsoft Orca demonstrated you can trade off size with larger fine-tuning dataset.
- tangjurine 3y agoAny references on this?
- treprinum 3y agohttps://www.youtube.com/watch?v=tVtOevLrt5U https://www.youtube.com/watch?v=tVtOevLrt5U