Hacker News Reader

Comment by stephc_int13 | Hacker News Reader

The current state of LLMs would be several orders of magnitude more impressive if they were only trained from data scrapped on the web.

But this is not the reality of modern LLMs by a long shot, they are trained in increasingly large parts from custom built datasets that are created by countless paid individuals, hidden behind stringent NDAs.

The author here seems to see that as a strength, an opportunity for unbounded growth and potential, I think this is the opposite, this approach is close to a gigantic whack a mole game, effectively unbounded, but in the wrong way.

View on HN

> countless paid individuals, hidden behind stringent NDAs.

If this is so prevalent, wouldn't there be a proportional amount of data leaks? If there any particular evidence, even of doubtful authenticity, of this being the case?

Those individuals just create data, they don't have access to it, think mechanical turk workers. All modern AI is powered by many such workers. LLM is the most funded modern AI, they have massive numbers of such workers for sure.

What sort of leak? I've seen data labeler/generation teams hired. I've never heard anyone describe the existence of these teams as a secret. No one hides the existence of Scale AI. People talk about which providers are better for different scenarios and when you need to inhouse and which companies are good at helping you build an inhouse team.

Are you talking about leaks of the actual training data? The secret sauce of modern LLMs? That is like leaking the google3 source code or the recipe for Coca Cola - a ludicrously risky move. And for what gain?

Can you give an example of these datasets and how they look like?

Do they ask PhD to explain root of negative 1 and why is it complex? Is it like Quora but private only and high quality answers by top researchers in their respective field?

Hmm, I don't want to talk specifics about my experience, but maybe check out some of the case studies on Surge's website - https://www.surgehq.ai/ (about halfway down the page).

> Do they ask PhD to explain root of negative 1 and why is it complex?

This isn't necessarily impossible, but I would consider it to be infeasible with existing labelling workforces. Of course if you really needed a dataset like this and you were sufficiently resourced and willing to spend, you could maybe make it work (I would question whether you really needed PhDs though, that might be hard to swing at any price point).

But the core idea behind your question is correct - this is what a dataset might look like and hiring/contracting appropriately-skilled people and asking them to do repetitive tasks with some guidance is how you would go about getting it. Depending on the need, it can be quite a bit more complex too - if you needed self-driving car driving behavior data maybe you build a simulator and hire people to drive in the simulator and use that as training data (made up and probably crap example, but it illustrates the possibilities).

Some people think that labelling workforces are all low skill and there is a lot of good things low skill workforces can do well (visual stuff, basic language and emotion tasks), but you might be surprised at the ability to get skilled labelers. There are lots of smart/educated people around the world and there is ridiculous amounts of money flowing into this space.

Yeah this is an interesting point. Other threads make the point about the "bitter lesson", and how expert-trained ML has historically not scaled, and human-generated LLM training data may just be repeating that dead end. Maybe so.

Something that is new this time around, AFAIK, is that we haven’t previously had general ML systems that businesses and consumers are paying billions of dollars a year to use. So if, say, 10% of revenue goes back in to making better data sets every year, I can imagine continued improvement on certain economically valuable use cases – though likely with diminishing returns.

Reminds me of the same issues with self driving. Seems like we need a completely different approach to solve these class of problems.

Reply on news.ycombinator.com