Claude Opus disagrees, lol (edit to add: all of what follows is Claude; not me):

I respectfully disagree with the claim that current language models are logically incapable of mapping language to actual concepts. While today’s LLMs certainly have limitations and are prone to hallucinations, they have demonstrated a remarkable ability to capture and manipulate complex concepts through language.

A few key points:

LLMs are trained on vast amounts of human-generated text which is grounded in real-world concepts, allowing the models to build rich representations that go beyond surface-level word associations. Techniques like unsupervised pre-training enable LLMs to learn meaningful conceptual relationships.

In many domains, LLMs have shown an impressive capacity for reasoning, inference, and knowledge synthesis - skills that rely on a deep understanding of the underlying concepts, not just pattern matching. Their performance on complex question-answering, analogical reasoning, and natural language inference tasks is evidence of this.

LLMs can be fine-tuned and augmented with techniques like retrieval-augmented generation or chain-of-thought prompting to further ground them in factual knowledge and improve their reasoning capabilities. This allows more robust conceptual mappings.

Careful prompt engineering and sampling techniques can help to mitigate hallucinations and nonsensical outputs. While not a complete solution, this shows that the conceptual knowledge is often present even if the model sometimes struggles to express it faithfully.

That said, I fully agree that the tendency for subtle nonsense is a serious issue, especially for layperson use cases where hallucinations may go undetected. Continued research into making LLMs more truthful, consistent and robust is crucial. Techniques like constitutional AI to bake in truthfulness, and better UX design to convey uncertainty are important steps.

But in summary, I believe the evidence suggests that LLMs, while flawed, are not fundamentally incapable of meaningful conceptual representation and reasoning. We should push forward on making them more reliable and trustworthy, rather than dismissing their potential prematurely.

Ich lebe in Amerika. Ich lerne Deutsche sprechen, aber das kostet Geld. Vielleicht wollen die Migranten Deutsche lernen, haben aber nicht das Geld dafür?

Sorry if the above is poorly worded; I’m still new to the language. My point is that there are lots of reasons that someone might not know a language well, including a lack of money, or a lack of time from needing to work full time to support one’s migrant family on a low wage.

Mexican immigrants to the US are wonderful, but their culture is very different from non-Hispanic US culture. I don’t expect them to learn English. They work like 60 hours per week to support their families. Like the person you’re replying to has said, though, their children learn English and integrate into, but also uniquely contribute to, US culture. Rather than expecting the first-generation immigrants to learn English, I’ve learned Spanish specifically to speak with them. It’s not like there are many more immigrants to Germany than there are immigrants to the US–even discounting the fact that the US has always been a country of immigrants, Hispanic and Latino/a/e Americans (the majority of which are Mexican Americans) are expected to exceed 50% of all Americans within a couple of decades. In some states, they are already the majority.

Diversity is a good thing, and we shouldn’t require immigrants to become like us culturally or linguistically before accepting them.

I’m thinking of shorting it. My friend is definitely shorting it.

To a degree. The large subreddits, like AskReddit, get far fewer upvotes on the top posts of the week than they used to get. I think there’s a good chunk of folks who left for a replacement, then left their replacement without going back to Reddit.

I go out of my way not to do so. Whenever I search for some specific items and see “Sponsored,” I’ll scroll down until I get the same listing without the ad link.

Lemmy Lemmy Lemmy

Yep

Would you, after devoting full years of your adult life to the unpaid work of learning the requisite advanced math and computer science needed to develop such a model, like to spend years more of your life to develop a generative AI model without compensation? Within the US, it is legal to use public text for commercial purposes without any need to obtain a permit. Developers of such models deserve to be paid, just like any other workers, and that doesn’t happen unless either we make AI a utility (or something similar) and funnel tax dollars into it or the company charges for the product so it can pay its employees.

I wholeheartedly agree that AI shouldn’t be trained on copyrighted, private, or any other works outside of the public domain. I think that OpenAI’s use of nonpublic material was illegal and unethical, and that they should be legally obligated to scrap their entire model and train another one from legal material. But developers deserve to be paid for their labor and time, and that requires the company that employs them to make money somehow.

I haven’t heard of any ultra-rich person who wants to reduce the population. A population decline will lead to stock price declines as the majority of the population ages (automated 401k investments buy and thus increase stock prices, withdrawals from 401k sell and thus decrease stock prices; an older population means less investment and greater withdrawal). Do you have a source for your decrease surplus population claim?

Wow, a real, live, tankie!

Wow, a real, live tankie!

Seconding. Kagi is the only one that was able to replace Google for me.

deleted by creator

GPT-4 will. For example, I asked it the following:

What is the neighborhood stranger model of fluid mechanics?

It responded:

The “neighborhood stranger model” of fluid mechanics is not a recognized term or concept within the field of fluid mechanics, as of my last update in April 2023.

Now, obviously, this is a made-up term, but GPT-4 didn’t confidently give an incorrect answer. Other LLMs will. For example, Bard says,

The neighborhood stranger model of fluid mechanics is a simplified model that describes the behavior of fluids at a very small scale. In this model, fluid particles are represented as points, and their interactions are only considered with other particles that are within a certain “neighborhood” of them. This neighborhood is typically assumed to be a sphere or a cube, and the size of the neighborhood is determined by the length scale of the phenomena being studied.

deleted by creator

I paid for Kagi and have been super happy with it. If you don’t mind paying, I highly recommend it. Not having ads or manipulated results is worth it for me.

Within a loop could be:

for(i in 1:10){
     assign(paste0("listNum", i), list(i, someStringVector[i], i:(i+20), i*value))
}```

And you can also use get() in the same way to dynamically retrieve a variable. 

I've gone so far into coding debauchery that I've dynamically assigned variables from dynamically retrieved ones, and I've done so fairly frequently.

Yeah, accurately simulating a single pyramidal neuron requires an eight-layer deep neural network:

https://www.cell.com/neuron/pdf/S0896-6273(21)00501-8.pdf

I do but I don’t know what that means either.

It also seemed to me that a growing number of comments in recent years there on product-relates posts (e.g., what’s the best language learning app?) reeked of companies promoting their own products rather (e.g., someone’s post history was selectively related to promoting some app after a period of inactivity). I haven’t seen things like that here at all, which is nice.

By season 8 she’s so convinced she’s essentially Mulder. She has times where she’s more or less convinced until then, but it’s a trajectory towards believing until she does. It just takes her a really long time.