andreyk

há 4 mesesdiscuss

"and we didn't see anything" is not justified at all.

Meta absolutely has (or at least had) a word class industry AI lab and has published a ton of great work and open source models (granted their LLM open source stuff failed to keep up with chinese models in 2024/2025 ; their other open source stuff for thins like segmentation don't get enough credit though). Yann's main role was Chief AI Scientist, not any sort of product role, and as far as I can tell he did a great job building up and leading a research group within Meta.

He deserved a lot of credit for pushing Meta to very open to publishing research and open sourcing models trained on large scale data.

Just as one example, Meta (together with NYU) just published "Beyond Language Modeling: An Exploration of Multimodal Pretraining" (https://arxiv.org/pdf/2603.03276) which has a ton of large-experiment backed insights.

Yann did seem to end up with a bit of an inflated ego, but I still consider him a great research lead. Context: I did a PhD focused on AI, and Meta's group had a similar pedigree as Google AI/Deepmind as far as places to go do an internship or go to after graduation.

andreyk

há 5 mesesdiscuss

This is quite misleading... From the article:

“When the Waymo vehicle encounters a particular situation on the road, the autonomous driver can reach out to a human fleet response agent for additional information to contextualize its environment,” the post reads. “The Waymo Driver [software] does not rely solely on the inputs it receives from the fleet response agent and it is in control of the vehicle at all times.” [from Waymo's own blog https://waymo.com/blog/2024/05/fleet-response/]

What's the problem with this?

andreyk

há 7 mesesdiscuss

To say they LLMs are 'predictive text models trained to match patterns in their data, statistical algorithms, not brains, not systems with “psychology” in any human sense.' is not entirely accurate. Classic LLMs like GPT 3 , sure. But LLM-powered chatbots (ChatGPT, Claude - which is what this article is really about) go through much more than just predict-next-token training (RLHF, presumably now reasoning training, who knows what else).

andreyk

há 7 mesesdiscuss

This blog post is full of bizarre statements and the author seems almost entirely ignorant of the history or present of AI. I think it's fair to argue there may be an AI bubble that will burst, but this blog post is plainly wrong in many ways.

Here's a few clarifications (sorry this is so long...):

"I should explain for anyone who hasn't heard that term [AI winter]... there was much hope, as there is now, but ultimately the technology stagnated. "

The term AI winter typically refers to a period of reduced funding for AI research/development, not the technology stagnating (the technology failing to deliver on expectations was the cause of the AI winter, not the definition of AI winter).

"[When GPT3 came out, pre-ChatGPT] People were saying that this meant that the AI winter was over, and a new era was beginning."

People tend to agree there were two AI winters already, one having to do with symbolic AI disappointments/general lack of progress (70s), and the latter related to expert systems (late 80s). That AI winter has long been over. The Deep Learning revolution started in ~2012, and by 2020 (GPT 3) huge amount of talent and money were already going into AI for years. This trend just accelerated with ChatGPT.

"[After symbolic AI] So then came transformers. Seemingly capable of true AI, or, at least, scaling to being good enough to be called true AI, with astonishing capabilities ... the huge research breakthrough was figuring out that, by starting with essentially random coefficients (weights and biases) in the linear algebra, and during training back-propagating errors, these weights and biases could eventually converge on something that worked."

Transformers came about in 2017. The first wave of excitement about neural nets and backpropagation goes all the way back to the late 80s/early 90s, and AI (computer vision, NLP, to a lesser extent robotics) were already heavily ML-based by the 2000s, just not neural-net based (this changed in roughly 2012).

"All transformers have a fundamental limitation, which can not be eliminated by scaling to larger models, more training data or better fine-tuning ... This is the root of the hallucination problem in transformers, and is unsolveable because hallucinating is all that transformers can do."

The 'highest number' token is not necessarily chosen, this depends on the decoding algorithm. That aside, 'the next token will be generated to match that bad choice' makes it sound like once you generate one 'wrong' token the rest of the output is also wrong. A token is a few characters, and need not 'poison' the rest of the output.

That aside, there are plenty of ways to 'recover' from starting to go down the wrong route. A key aspect of why reasoning in LLMs works well is that it typically incorporates backtracking - going earlier in the reasoning to verify details or whatnot. You can do uncertainty estimation in the decoding algorithm, use a secondary model, plenty of things (here is a detailed survey https://arxiv.org/pdf/2311.05232 , one of several that is easy to find).

"The technology won't disappear – existing models, particularly in the open source domain, will still be available, and will still be used, but expect a few 'killer app' use cases to remain, with the rest falling away."

A quick google search shows ChatGPT currently has 800 million weekly active users who are using it for all sorts of things. AI-assisted programming is certainly here to stay, and there are plenty of other industries in which AI will be part of the workflow (helping do research, take notes, summarize, build presentations, etc.)

I think discussion is good, but it's disappointing to see stuff with this level of accuracy being on front page of HN.

andreyk

há 9 mesesdiscuss

For reference, the details about how the LLMs are queried:

"How the players work

    All players use the same system prompt
    Each time it's their turn, or after a hand ends (to write a note), we query the LLM
    At each decision point, the LLM sees:
        General hand info — player positions, stacks, hero's cards
        Player stats across the tournament (VPIP, PFR, 3bet, etc.)
        Notes hero has written about other players in past hands
    From the LLM, we expect:
        Reasoning about the decision
        The action to take (executed in the poker engine)
        A reasoning summary for the live viewer interface
    Models have a maximum token limit for reasoning
    If there's a problem with the response (timeout, invalid output), the fallback action is fold"

The fact the models are given stats about the other models is rather disappointing to me, makes it less interesting. Would be curious how this would go if the models had to only use notes/context would be more interesting. Maybe it's a way to save on costs, this could get expensive...

andreyk

há 9 mesesdiscuss

But LLMs would presumably also condition on past observations of opponents - i.e. LLMs can conversely adapt their strategy during repeated play (especially if given a budget for reasoning as opposed to direct sampling from their output distributions).

The rules state the LLMs do get "Notes hero has written about other players in past hands" and "Models have a maximum token limit for reasoning" , so the outcome might be at least more interesting as a result.

The top models on the leaderboard are notably also the ones strongest in reasoning. They even show the models' notes, e.g. Grok on Claude: "About: claude Called preflop open and flop bet in multiway pot but folded to turn donk bet after checking, suggesting a passive postflop style that folds to aggression on later streets."

PS The sampling params also matter a lot (with temperature 0 the LLMs are going to be very consistent, going higher they could get more 'creative').

PPS the models getting statistics about other models' behavior seems kind of like cheating, they rely on it heavily, e.g. 'I flopped middle pair (tens) on a paired board (9s-Th-9d) against LLAMA, a loose passive player (64.5% VPIP, only 29.5% PFR)'

andreyk

há 4 anosdiscuss

Yep. The people who build Imagen are researchers, not engineers, and these announcements are accompanied by papers describing the results as a means of sharing ideas/results with the academic community. Pretty weird to me how so many in this thread don't seem to remember that.

andreyk

há 7 anosdiscuss

Boils down to this: "So what is the role of architecture patterns? I see them similarly in usefulness as coding design patterns. They can give you ideas on how to improve your code or architecture."

The whole idea of patterns is to identify often useful, and possibly non-obvious, ideas to be aware of when designing the solution. It's great to start simple, but tricky to make things both simple and robust/powerful - and that's what patterns are supposed to help with. This ends with:

"Software architecture best practices, enterprise architecture patterns, and formalized ways to describe systems are all tools that are useful to know of and might come in handy one day. But when designing systems, start simple and stay as simple as you can. Try to avoid the complexity that more complex architecture and formal tools inherently introduce."

What this misses that if you start simple and stay as simple as you can, you may undershoot and be stuck refactoring code down the line; a fine balance is needed, and patterns are definitely part of a toolset that a good engineer should be aware of when trying to nail that balance.

Submissions

Lurie promised a permitting overhaul. It was troubled from the start

Notes from an Authoritarian Year

comments