Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Notes on OpenAI's new o1 chain-of-thought models

simonwillison.net

Post reply on HN
Print view View original

151–160 of 659 posts

Re: Notes on OpenAI's new o1 chain-of-thought models

#151

Post by fragmede »

pedrosorio wrote: post #110
MPSimmons wrote: post #106

Earlier quoted context omitted.

The failure is in how you're using it. I don't mean this as a personal attack, but more to shed light on what's happening. A lot of people use LLMs as a search engine. It makes sense - it's basically a lossy compressed database of everything its ever read, and it generates output that is statistically likely - varying degrees of likeliness depending on the temperature, as well as how many times the particular weights…

> It knows english at or above a level equal to most fluent speakers, and it also can produce output that is not just a likely output, but is a logical output This is not an apt description of the system that insists the doctor is the mother of the boy involved in a car accident when elementary understanding of English and very little logic show that answer to be obviously wrong. https://x.com/colin_fraser/status/183…

what's weird is it gets it right when I try it.

https://chatgpt.com/share/66e3601f-4bec-8009-ac0c-57bfa4f059...

Re: Notes on OpenAI's new o1 chain-of-thought models

#152

Post by ksynwa »

> first introduced in the paper Large Language Models are Zero-Shot Reasoners in May 2022

What's a zero shot reasoner? I googled it and all the results are this paper itself. There is a wikipedia article on zero shot learning but I cannot recontextualise it to LLMs.

Re: Notes on OpenAI's new o1 chain-of-thought models

#153

Post by re-thc »

Kiro wrote: post #148
re-thc wrote: post #87

Earlier quoted context omitted.

> I'm not writing that much of my own code at my FAANG job. > Most of the time, I just accept its changes. This speaks more about the problems at FAANG, other companies, etc than AI vs a human developer. And AI isn't the real fix. Are we just repeating things 100x a day or is it still so chaotic and immature? Or are we implying that AI is at a point where it's writing Google Spanner from scratch and you're able to re…

We went from "it's useless because..." - "it outputs gibberish" to "it just copypastes" to "it only works for simple things" to "it can't make Google Spanner from scratch".

> We went from

None of the above.

This isn't about how "smart" AI is.

1. Let's assume it was smart and can update a field spanning 1000s of microservices to deliver this new feature. Is this really something you should celebrate? I'd say no. At this point there should have been better tooling and infrastructure in place.

2. Is there really infinite CRUD to add after >10 years? In the same organization where you need >100s of developers all the time? 1s where you'd ignore code reviews and "just accept its changes"? Whether I write code or my colleagues etc I'd have a meaningful discussion about the proposed changes, the impacts and most likely suggest changes because nothing is perfect.

So again, it's about the environment, the organization or at least this individual case where coding isn't just about adding some lines to a file. And that's with AI or not.

Re: Notes on OpenAI's new o1 chain-of-thought models

#154

Post by thatguymike »

ksynwa wrote: post #152

> first introduced in the paper Large Language Models are Zero-Shot Reasoners in May 2022 What's a zero shot reasoner? I googled it and all the results are this paper itself. There is a wikipedia article on zero shot learning but I cannot recontextualise it to LLMs.

It used to be that you had to give examples of solving similar problems to coax the LLM to solve the problem you wanted it to solve, like: """ 1 + 1 = 2 | 92 + 41 = 133 | 14 + 6 = 20 | 9 + 2 = """ -- that would be an example of 3-shot prompting.

With modern LLMs you still usually get a benefit from N-shot. But you can now do "0-shot" which is "just ask the model the question you want answered".

Re: Notes on OpenAI's new o1 chain-of-thought models

#155

Post by fsndz »

layer8 wrote: post #18

The o1-preview model still hallucinates non-existing libraries and functions for me, and is quickly wrong about facts that aren't well-represented on the web. It's the usual string of "You're absolutely correct, and I apologize for the oversight in my previous response. [Let me make another guess.]" While the reasoning may have been improved, this doesn't solve the problem of the model having no way to assess if what…

My point of view: this is a real advancement. I've always believed that with the right data allowing the LLM to be trained to imitate reasoning, it's possible to improve its performance. However, this is still pattern matching, and I suspect that this approach may not be very effective for creating true generalization. As a result, once o1 becomes generally available, we will likely notice the persistent hallucinations and faulty reasoning, especially when the problem is sufficiently new or complex, beyond the "reasoning programs" or "reasoning patterns" the model learned during the reinforcement learning phase. https://www.lycee.ai/blog/openai-o1-release-agi-reasoning

Re: Notes on OpenAI's new o1 chain-of-thought models

#156

Post by simonw »

dkga wrote: post #143

Judging by this thread, surely a part of OpenAI’s business model is to release models with somewhat grey-area outlandish claims, then sit and wait for people to test it out paying top dollars for tokens.

They’re not charging anything for o1 right now - it’s running as a free beta.

EDIT: Actually I’m not sure I’m correct about that - I’ve seen screenshots that indicate the API model is being charged for here: https://twitter.com/deepaiwriter/status/1834460643228897512

The value OpenAI get here is that people effectively run a massively parallel brute force attack against the new models to figure out exactly what they can and can’t do.

Re: Notes on OpenAI's new o1 chain-of-thought models

#157

Post by janalsncm »

ChicagoDave wrote: post #137

It’s still just a tool. It does not reason. It has some add-on logic the simulates it. We’re no closer to “AI” today than we were 20 years ago.

> We’re no closer to “AI” today than we were 20 years ago.

20 years ago we had barely figured out how to create superhuman agents to play chess. We have now created a new algorithm to solve Go, which is a much harder game.

We then created an algorithm (alpha zero) to teach itself to play any game, and which became the best chess player in the world in hours.

We next created a superhuman poker agent. Poker is even more complex than Go because it involves imperfect information and opponent modeling.

We then created a superhuman agent to play Diplomacy, which requires natural language and cooperation with other humans to reason about imperfect (hidden) information.

Re: Notes on OpenAI's new o1 chain-of-thought models

#159

Post by koe123 »

mattmanser wrote: post #150
COAGULOPATH wrote: post #54

Earlier quoted context omitted.

Yes, this only helps multi-step reasoning. The model still has problems with general knowledge and deep facts. There's no way you can "reason" a correct answer to "list the tracklisting of some obscure 1991 demo by a band not on Wikipedia." You either know or you don't. I usually test new models with questions like "what are the levels in [semi-famous PC game from the 90s]?" The release version of GPT-4 could get abo…

It's actually much worse than that and you're inadvertently down playing how bad it is. It doesn't even know mildly obsecure facts that are on the internet. For example last night I was trying to do something with C# generics and it confidently told me I could use pattern matching on the type in a switch statwmnt, and threw out some convincing looking code. You can't, it's impossible. It wàa completely wrong. When I…

Its not always the right tool depending on the task. IMO using LLMs is also a skill, much like learning how to Google stuff.

E.g. apparently C# generics isn’t something its good at. Interesting, so don’t use it for that, apparently its the wrong tool. In contrast, its amazing at C++ generics, and thus speeds up my productivity. So do use it for that!

Re: Notes on OpenAI's new o1 chain-of-thought models

#160

Post by simonw »

ChicagoDave wrote: post #137

It’s still just a tool. It does not reason. It has some add-on logic the simulates it. We’re no closer to “AI” today than we were 20 years ago.

Personally I think “add-on logic that simulates reasoning” is a pretty good match for the “artificial” part of “artificial intelligence”.

I’ve been tryin out the alternative term “initiation intelligence” recently, mainly to work around the baggage that’s become attached to the term AI.

Post reply on HN