Rudybega

7 ms·

31.
by Rudybega 1y ago
I think it's probably still useful to distinguish addictions with hardcore substance related barriers to quitting (think withdrawals) from addictions where the barrier is a lack of dopamine or serotonin or simple habituation. For peopl
32.
by Rudybega 1y ago
Scaffolding typically just refers to a larger state machine style control flow governing an agent's behavior and the suite of external tools it has access to.
33.
by Rudybega 1y ago
This is actively being worked on my pretty much every major provider. It was the subject of that recent OpenAI paper on hallucinations. It's mostly caused by benchmarks that reward correct answers, but don't penalize bad answers
34.
by Rudybega 1y ago
Yeah, this is a pretty major error. They already have a higher revenue than the article's quoted "jump next year" (which was this year and was an underestimate).
35.
by Rudybega 1y ago
> what about the fact frontier labs are spending more compute on viral AI video slop and soon-to-be-obsoleted workplace usecases than research? That's a bold claim, please cite your sources. It's hard to find super precise sour
36.
by Rudybega 1y ago
I wonder if you could dramatically improve these results with some relatively simple scaffolding and tool access. If a ton of these mistakes are genuinely simple calculation errors, it seems like giving the models access to a calculator too
37.
by Rudybega 1y ago
Higher token throughput is great for use cases where the smaller, faster model still generates acceptable results. Final response time improvements feel so good in any sort of user interface.
38.
by Rudybega 1y ago
> The largest model had 20x more clean data than dirty data in the training data. Yeah, I think this is the main misinterpretation. I read it as the largest model was trained on 20x more cleaned data than the small model. I don't t
39.
by Rudybega 1y ago
I'd love to see a blind study comparing a wide spectrum of these AI tracks to lesser known real artists (so the participants don't just recognize the songs) to see whether people can tell or if knowledge of the source biases them.
40.
by Rudybega 1y ago
I actually think the vocals from ~2:00-~2:35 are pretty impressive there. It's wild to me that the models can play with tempo like that. I've been listening to this across a variety of genres though, maybe these lyrics and vocals
41.
by Rudybega 1y ago
I think AI is starting to verge on making actual good music. The latest Suno release is wild. An example here: https://v.redd.it/fqlqrgumo5rf1 I find this one interesting because Rap has classically been difficult for thes
42.
by Rudybega 1y ago
Sorry, those figures are skewed by Timelord Georg, who has been using AI for 100 million hours a day, is an outlier, and should have been removed.
43.
by Rudybega 1y ago
If we develop a system that can: - compress (in a relatively recoverable way) the entire domain of human knowledge - interpolate across the entire domain of human knowledge - draw connections or conclusions that haven't previously been
44.
An AI system to help scientists write expert-level empirical software(arxiv.org)
3 points by Rudybega 1y ago|1 comments
45.
by Rudybega 1y ago
Abstract: "The cycle of scientific discovery is frequently bottlenecked by the slow, manual creation of software to support computational experiments. To address this, we present an AI system that creates expert-level scientific softwa
46.
by Rudybega 1y ago
This is true, but there are methods to greatly reduce the effect of this and generate results that match or even improve overall output accuracy: e.g. DOMINO https://arxiv.org/html/2403.06988v1
47.
by Rudybega 1y ago
This is what they tell you in driver's education to try and reduce the odds you hit pedestrians, but it's not legally true in most jurisdictions.
48.
by Rudybega 1y ago
This mostly gets avoided with maximum buyin limits on tables. Sure someone can accrue well over that amount, but that's part of cash poker too (dealing with players with large stacks). Obviously players with much larger bankrolls can
49.
by Rudybega 1y ago
You are correct. I need some coffee.
50.
by Rudybega 1y ago
I think OpenAI has projected 12.7B in revenue this year and 29.4B in 2026. Edit: I am dumb, ignore the second half of my post.
51.
by Rudybega 1y ago
It's not really a function of maintaining coherency across context length. It's more about whether the model can accomplish a long time horizon task when the context length of a single message isn't even close to sufficient
52.
by Rudybega 1y ago
All of that is totally unrelated to the point I'm trying to make. Pokemon is interesting because it's a test of whether these models can solve long time horizon tasks. That's it.
53.
by Rudybega 1y ago
Shh, don't worry and just embrace the spiral. Edit: No spiral emojis allowed, clearly this site will be the first to fall.
54.
by Rudybega 1y ago
He has spoken repeatedly about running for a third term.
55.
by Rudybega 1y ago
The state space for actions in Pokemon is hilariously, unbelievably larger than the state space for chess. Older chess algorithms mostly used Brute Force (things like minimax) and the number of actions needed to determine a reward (winning
56.
by Rudybega 1y ago
I mostly use Roo Code inside visual studio. The modes are awesome for managing context length within a discrete unit of work.
57.
by Rudybega 1y ago
I'm going to have to heavily disagree. Gemini 2.5 Pro has super impressive performance on large context problems. I routinely drive it up to 4-500k tokens in my coding agent. It's the only model where that much context produces
58.
by Rudybega 1y ago
You could just use something like roo code with custom modes rather than manually injecting them. The orchestrator mode can decide on the other appropriate modes to use for subtasks. You can customize the system prompts, baseline propmts,
59.
by Rudybega 1y ago
The context7 mcp works similarly. It allows you to search a massive constantly updated database of relevant documentation for thousands of projects.
60.
by Rudybega 1y ago
This is kind of what the modes in roo code do now. I'm having great success with them and having them as a default just rolled out a couple days ago. There are a default set of modes (orchestrator, code, architect, debug, and ask) and

More ›