red75prime

12 ms·

331.
by red75prime 8mo ago
> there are significant limitations Where can we read about those significant limitations?
332.
by red75prime 8mo ago
I think the relevant question is: can a statistical model (or a transformer, in particular) generalize to general reasoning ability?
333.
by red75prime 8mo ago
An intelligent mind should keep a balance. U: The car wash is 50m away. Should I walk or should I drive? A: Do you need to wash your car? U: Yep. A: Is the car yours? U: Yep. A: Is your car nearby? U: Yes. A: Do you need to walk to your car
334.
by red75prime 8mo ago
I wasn't talking about free speech per se, but about the fragmentation of public discourse. Having multiple popular platforms makes fragmentation easier. All your friends are here. All nazies/wokes are there. There's nothing
335.
by red75prime 8mo ago
Yeah, this way there's no need to debunk egregious lies or listen to the other side's replies. You have a feed of debunked lies at home.
336.
by red75prime 8mo ago
More choice, more competition, more echochambers.
337.
by red75prime 8mo ago
It’s more like, "Some people have something going on that ameliorates the cardiovascular and cancer problems caused by smoking."
338.
by red75prime 8mo ago
Obviously, there’s a limit to how much you can squeeze into a single parameter. I guess the low-hanging fruit will be picked up soon, and scaling will continue with algorithmic improvements in training, like [1], to keep the training comput
339.
by red75prime 8mo ago
The former doesn't prevent tens of thousands deaths each year.
340.
by red75prime 8mo ago
There's no evidence for that. And there's no sense for Tesla doing it. Politics is really a mind killer. Just think for a second. Who can be fooled by this "turning off FSD milliseconds before impact"?
341.
by red75prime 8mo ago
No info on that. 13781-13647 Street, Other fixed object, No injuries, Proceeding Straight, 17mph, contact area: bottom 13781-13648 Street, Bus, No injuries, Stopped, 0mph, contact area: left, front 13781-13646 Parking lot, Other fixed objec
342.
by red75prime 8mo ago
> I find it more plausible that the feeling of speaking from the gut is akin to allowing that center of intelligence coordinate the rest of our body. The author thinks that mentally poking the bodymap created by the brain can cause chang
343.
by red75prime 8mo ago
I mean reasoning models don't seem to make this mistake (so, System 1) and the mistake is not universal across models, so a "hiccup" (a brain hiccup, to be precise).
344.
by red75prime 8mo ago
A hiccup in a System 1 response. In humans they are fixed with the speed of discovery. Continual learning FTW.
345.
by red75prime 8mo ago
All the latest general purpose models are multimodal (except DeepSeek I think). Transfer learning allows to improve results even after they exhausted all the text in the internet.
346.
by red75prime 8mo ago
Python is not a natural language. A natural language communicates things that are not here and now (essentially everything that occurs naturally). Moreover, language can convey abstractions that exist because of language itself.
347.
by red75prime 8mo ago
Aren't they from somewhat below waist level a parte posteriori ?
348.
by red75prime 8mo ago
I wonder if psychology plays a role here. An engineer with strong manual coding skills might be hesitant to admit that a tool has become good enough to warrant less involvement.
349.
by red75prime 8mo ago
> Waymo self driving never disengages the way Tesla FSD does. It is active at all times Consumer version of FSD can park a car if a driver doesn't take contol[1]. Waymo seems to require a remote command to initiate parking instead o
350.
by red75prime 8mo ago
If a person has motivation, it's not impossible to find means. Hitchhiking. General assistance thru the Human Services Agency for cash. Explain your circumstances/try another job offering/repeat.
351.
by red75prime 8mo ago
Evolution has scaled the brain to hundreds of trillions of synapses (parameters) and hundreds of billions of neurons (processing nodes). Current models have around 1/1000 of that number of parameters. The top data centers operate at ar
352.
by red75prime 8mo ago
Waymo routinely uses safety drivers, sorry, "autonomous specialists" when expanding to new cities[1][2]. Waymo cars occasionally contact the remote support. If support is not available, the cars just stay where they stopped[3]. Te
353.
by red75prime 8mo ago
Sure. Language just seems to structure a general learning system in a way that allows it to do all those things.
354.
by red75prime 8mo ago
But the real intelligence exists and it is beyond all those trinkets. Right? Yeah, this is a common cope. There's no solid evidence that the brain uses quantum computations. And that's the only definite thing (besides 'magic&
355.
by red75prime 8mo ago
[0] if you want something more academic. What's your gripe with language, anyway? [0] https://arxiv.org/abs/2506.24044
356.
by red75prime 8mo ago
Language allows higher-level and longer-term planning and better generalization than purely vision-action models. Waymo uses VLM in their Driver[1]. Tesla decided to add some kind of a language model to their VA stack[2]. Nvidia's Alpa
357.
by red75prime 8mo ago
The article feels, I don't know… maybe like someone calmly sitting in a rocking chair staring at the sea. Then the camera turns, and there's an erupting volcano in the background. > If it was a life or death decision, would you
358.
by red75prime 8mo ago
"Sweat and tears" -> exploration and the training signal for reinforcement learning.
359.
by red75prime 8mo ago
It's relatively easy to analyze skill files. Shared chunks of neural networks (LLM adapters) can hide malicious behaviors better.
360.
by red75prime 8mo ago
Read Electrek articles with a mouthful of salt. Fred Lambert’s “robotaxi is 10x worse than a human” estimate is based on his personal statistical reasoning, which somehow arrived at 200,000 miles per accident for humans. Minor accidents tha

More ›