fxwin

I could agree with that, but the example on p29 (Claim 6) ends with QED, but only then the proof follows. I realize I'm nitpicking form here, but still

Hah, that's pretty clever if it's true .D

Seems like this should be in the paper! Thanks though

could just be non-english autocorrect or keyboard layout

I haven't read through the entire thing yet, but the long abstract combined with the way the acronym BDH is introduced (What does the B stand for?) along with the very "flowery" name (When neither "dragon" nor "hatchling" appears again past page 2) is rather offputting

- It seems strange to make use of the term "scale-free" and then defer a definition until half way through the paper (in fact, the term is mentioned 3 times after, and 14 times before said definition)

- This might just be CS people doing CS things, but the notation in the paper is awful: Claims/Observations end with a QED-symbol (for example on pages 29 and 30) but without a proof

- They make strong claims about performance and scaling ("It exhibits Transformer-like scaling laws") but the only (i think?) benchmark is a translation task comparison with <1B models, ,which is ~2 orders of magnitude smaller than sota

German newspaper "Die Zeit" has a few videos where they get art/cultural critics to watch and comment on modern meme culture which i find quite entertaining. Here is the video about Skibidi Toilet: https://youtu.be/z-oAtxjnDlQ?si=FjpcVJxMoLv537RZ (Audio is mostly german, but the subtitles are quite accurate from what i can tell).

Also recommend this one with "German Brainrot": https://www.youtube.com/watch?v=-mJENuEN_rs

Thanks, I didn't see that part!

> this is about using vision-LLMs for OCR

Is it? To me it sounds like they do OCR first, then extract from the result with LLM:

"Cortex uses automated extraction (optical character recognition (OCR) and natural language processing (NLP)) to parse clerkship grades from medical school transcripts."

I would assume they OCR first, then extract whatever info they need from the result using LLMs

Edit: Does sound like it - "Cortex uses automated extraction (optical character recognition (OCR) and natural language processing (NLP)) to parse clerkship grades from medical school transcripts."

I would agree with that

> held out

i think this is what you're looking for: https://www.media.mit.edu/projects/alterego/publications/

Same here. I've been working through some textbooks without solutions for the contained exercises, and ChatGPT has been invaluable for getting feedback on solutions and hints when I'm stuck

not to anyone? really?

https://i.imgur.com/cf1wOwL.png after a few minutes of running it at 240 frames per second :D

you can run frameRate(x) in your browser's console to speed it up a bit - might be limited by your monitor's refresh rate though

This makes no sense, the comment obviously implies that there are in fact serious orgs using twitter

that's why he said it describes how runtime behaves as input size changes, not how it behaves for specific values

Also known as the Combed Hedgehog Theorem (which i like a bit better)

The design in the OP is still more appealing (if less practical) to me because its movements and how it plays are still fairly close to how a human would play (Precisely because of the sliding). Sort of like how we generally find it less interesting when a specialized robot can perform some specific task rather than say a humanoid robot, even if it is worse at said task.

I'm also not the biggest fan of the white-on-black style, but there is definitely precedent (at least in science-youtube-space) for explainer videos "drawn live" [1-4]

[1] https://www.youtube.com/@Aleph0

[2] https://www.youtube.com/@MinutePhysics

[3] https://www.youtube.com/@12tone

[4] https://www.youtube.com/@SimplilearnOfficial

Ah, you are right! At least wrt the "naive" mapping of irrationals to binary representation. My gut still tells me it's bijective, but the mapping has to be a bit more involved :D

> One could then wonder which irrational numbers are represented by such games.

Seems like there should be a bijection there no?

it doesn't get much easier than that, and the downsides are much much much less of an inconvenience than having your data breached depending on what it is.

could also be just massively scaling up energy consumption with little concern for efficiency (since limitless would imply very low cost), which would probably be a bad idea for renewables, and in case of not-so-cheap energy also very expensive

all of which are costs that pale in comparison to having your data compromised, depending on what it is

You might be right that we'll never have quantum computers capable of cracking conventional cryptographic methods, but I'd rather err on the side of caution in this regard considering how easy it is to switch, and how disastrous it could be otherwise.

> People claimed all the basics were done, that we "only" needed to scale.

This claim is fundamentally different from what you quoted.

> But who's prepping for unlimited energy today?

It's about tradoffs: It costs almost nothing to switch to PQC methods, but i can't see a way to "prep for unlimited energy" that doesn't come with huge cost/time-waste in the case that doesn't happen

The page only talks about adopting PQC for key agreement for SSH connections, not encryption in general so the overhead would be rather minimal here. Also from the FAQ:

"Quantum computers don't exist yet, why go to all this trouble?"

Because of the "store now, decrypt later" attack mentioned above. Traffic sent today is at risk of decryption unless post-quantum key agreement is used.

"I don't believe we'll ever get quantum computers. This is a waste of time"

Some people consider the task of scaling existing quantum computers up to the point where they can tackle cryptographic problems to be practically insurmountable. This is a possibilty. However, it appears that most of the barriers to a cryptographically-relevant quantum computer are engineering challenges rather than underlying physics. If we're right about quantum computers being practical, then we will have protected vast quantities of user data. If we're wrong about it, then all we'll have done is moved to cryptographic algorithms with stronger mathematical underpinnings.

Not sure if I'd take the cited paper (while fun to read) too seriously to inform my opinion the risks of using quantum-insecure encryption rather than as a cynical take on hype and window dressing in QC research.

Then the "Examples" section should include a "How to build a simple MCP server using this framework" section. Instead, it contains an example for an "MCP client" (however the implementation has little to do with what the MCP spec says a client does), and an "example of a custom module for the MCP server" (which MCP server? we haven't built one yet), which (after checking implementation) does not conform to the MCP spec for tools (see https://modelcontextprotocol.io/specification/2025-06-18/ser...).

Unless I'm missing something, this has very little to nothing to do with MCP.