401–410 of 643 posts
Indeed. I don't understand why Hacker News is so dismissive about the coming of LLMs, maybe HN readers are going through 5 stages of grief? But LLM is certainly a game changer, I can see it delivering impact bigger than the internet itself. Both require a lot of investments.
So we are just irrational and sour?
Speaking of asynchronous agents, what do people use? Claude Code for web is extremely limited, because you have no custom tools. Claude Code in GitHub Actions is vastly more useful, due to the custom environment, but ackward to use interactively. Are there any good alternatives?
I'm running Claude Code in a tmux on a VPS, and I'm working on setting up a meta-agent who can talk to me over text messages
Would you be open to providing more details. Would love to hear more, your workflows, etc.
Earlier quoted context omitted.
Robotics is coming FAST. Faster than LLM progress in my opinion.
The question is how rapid the adoption is. The price of failure in the real world is much higher ($$$, environmental, physical risks) vs just "rebuild/regenerate" in the digital realm.
Earlier quoted context omitted.
Robotics is coming FAST. Faster than LLM progress in my opinion.
Curious if you have any links about the rapid progression of robotics (as someone who is not educated on the topic). It was my feeling with robotics that the more challenging aspect will be making them economically viable rather than simply the challenge of the task itself.
Earlier quoted context omitted.
No idea about training tenserflow models - is it super complex or is it just calling a couple of APIs ? Langchain is literally calling an API. Maybe you need to get good with prompting or whatever, but I don't see where the complexity lies. Please let me know.
Having used both Tensorflow (though I expect they mean PyTorch which is way more popular, and I have also used) and langchain, they are nothing alike. They he ML frameworks are much closer to implementing the mathematics of neural networks, with some abstractions but much closer to the linear algebra level. It requires an understanding of the underlying theory. Langchain is a suite of convenience functions for compos…
Building the model may range from very simple if you are just recreating a standard architecture, or be a research endeavor if you are designing something completely new.
The difficulty/complexity of then training the model depends on what it is. For something simple like a CNN for image recognition, it's really just a matter of selecting a few hyperparameters and letting it rip. At the other end of the spectrum you've got LLMs where training (and coping with instabilities) is something of a black art, with RL training completely different from pre-training, and there is also the issue of designing/discovering a pre/mid/post training curriculum.
But anyways, the actual training part can be very simple, not requiring too much knowledge of what's going on under the hood, depending on the model.
2025: The Year in Open Source? Nothing, all resources were tied up to debunk a couple of Python web developers who pose as the ultimate experts in LLMs.
Earlier quoted context omitted.
The emergent phenomenon is that the LLM can separate truth from fiction when you give it a massive amount of data. It can figure the world out just as we can figure it out when we are as well inundated with bullshit data. The pathways exist in the LLM but it won’t necessarily reveal that to you unless you tune it with RL.
> The emergent phenomenon is that the LLM can separate truth from fiction when you give it a massive amount of data. I don't believe they can. LLMs have no concept of truth. What's likely is that the "truth" for many subjects is represented way more than fiction and when there is objective truth it's consistently represented in similar way. On the other hand there are many variations of "fiction" for the same subject…
So think of it like this, to create the model we use terabytes of data. Then we do RL which is probably less than one percent of additional data involved in the initial training.
The change in the model is that reliability is increased and hallucinations are reduced at a far greater rate than one percent. So much so that modern models can be used for agentic tasks.
How can less than one percent of reinforcement training get the model to tell the truth greater than one percent of the time?
The answer is obvious. It ALREADY knew the truth. There’s no other logical way to explain this. The LLM in its original state just predicts text but it doesn’t care about truth or the kind of answer you want. With a little bit of reinforcement it suddenly does much better.
It’s not a perfect process and reinforcement learning often causes the model to be deceptive an not necessarily tell the truth but it more gives an answer that may seem like the truth or an answer that the trainer wants to hear. In general though we can measurably see a difference in truthfulness and reliability to an extent far greater than the data involved in training and that is logical proof it knows the difference.
Additionally while I say it knows the truth already this is likely more of a blurry line. Even humans don’t fully know the truth so my claim here is that an LLM knows the truth to a certain extent. It can be wildly off for certain things but in general it knows and this “knowing” has to be coaxed out of the model through RL.
Keep in mind the LLM is just auto trained on reams and reams of data. That training is massive. Reinforcement training is done on a human basis. A human must rate the answers so it is significantly less.
Earlier quoted context omitted.
By that definition, nearly all commercial software development (and nearly all human output in general) is derived output.
Wow. You’re using ‘derived’ to imply ‘therefore equivalent.’ That’s a category error. A cookbook is derived from food culture. Does an LLM taste food? Can it think about how good that cookie tastes? A flight simulator is derived from aerodynamics - yet it doesn’t fly. Likewise, text that resembles reasoning isn’t the same thing as a system that has beliefs, intentions, or understanding. Humans do. LLMs don't. Also...…
I imagine that sounded more profound when you wrote it than it did just now, when I read it. Can you be a little more specific, with regard to what features you would expect to differ between LLM and human responses to such a question?
Right now, LLM system prompts are strongly geared towards not claiming that they are humans or simulations of humans. If your point is that a hypothetical "thinking" LLM would claim to be a human, that could certainly be arranged with an appropriate system prompt. You wouldn't know whether you were talking to an LLM or a human -- just as you don't now -- but nothing would be proved either way. That's ultimately why the Turing test is a poor metric.
Earlier quoted context omitted.
By that definition, nearly all commercial software development (and nearly all human output in general) is derived output.
Wow. You’re using ‘derived’ to imply ‘therefore equivalent.’ That’s a category error. A cookbook is derived from food culture. Does an LLM taste food? Can it think about how good that cookie tastes? A flight simulator is derived from aerodynamics - yet it doesn’t fly. Likewise, text that resembles reasoning isn’t the same thing as a system that has beliefs, intentions, or understanding. Humans do. LLMs don't. Also...…