frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Sonnet 4.6 model could mistakenly use wrong model for OpenAI

https://github.com/anthropics/claude-code/issues/51417
1•keytalker•37s ago•0 comments

What Is Multi-Cloud Security? Challenges and Best Practices

https://spacelift.io/blog/multi-cloud-security
1•kat-w•1m ago•0 comments

RLMs are the new reasoning models

https://raw.works/rlms-are-the-new-reasoning-models/
1•dnw•12m ago•0 comments

Gell-Mann AImnesia

https://huonw.github.io/blog/2026/04/gell-mann-aimnesia/
1•dbaupp•13m ago•1 comments

Mall Code

https://mall.merkoba.com
1•madprops•15m ago•0 comments

Anthropic says OpenClaw-style Claude CLI usage is allowed again

https://docs.openclaw.ai/providers/anthropic
2•jmsflknr•17m ago•0 comments

No Naked Singularity, Whatever the Physical Collapse

https://zenodo.org/records/16181570
2•jruohonen•17m ago•0 comments

Svelte-check-native: Blazing fast svelte-check built with Rust

https://github.com/harshmandan/svelte-check-native
1•thunderbong•20m ago•0 comments

US Utilities Plan $1.4T for AI Data Centers

https://tech-insider.org/us-utility-1-4-trillion-ai-data-center-energy-2026/
2•jackyli02•20m ago•0 comments

Theseus, a Static Windows Emulator

https://neugierig.org/software/blog/2026/04/theseus.html
1•matt_d•20m ago•0 comments

PageGuard – scan a URL, get compliance docs from the actual tech detected

https://www.getpageguard.com
1•Dhicks_builds•20m ago•0 comments

Smart Home for Beginners: Where to Start

https://aigadgetexpert.com/best-smart-home-beginners-2026
1•amghal•22m ago•0 comments

Jersey Mike's confidentially files for IPO

https://www.cnbc.com/2026/04/20/jersey-mikes-ipo.html
1•lxm•23m ago•0 comments

Pica: Better Font Management for macOS

https://pica.joshpuckett.me/
2•jbegley•26m ago•0 comments

Stb_AVIF: A pure C89, Libc-only AVIF decoder in stb-style single-header form

https://github.com/lenchan139/stb_avif
1•roytam87•28m ago•1 comments

Can you make a picture of a dog wearing a hat

https://dispatchesfromthefuture.substack.com/p/can-you-make-a-picture-of-a-dog-wearing
1•JoiDegn•28m ago•0 comments

Show HN: Local, agent-friendly double-entry bookkeeping and tax prep

https://github.com/andrewchilds/moneypit
1•andrewchilds•29m ago•0 comments

Substack added a scheduler. Here's why I kept building PubQ anyway

https://www.indiehackers.com/post/substack-added-a-scheduler-heres-why-i-kept-building-pubq-anywa...
1•rkapdi•30m ago•0 comments

Trump's Landman Iran Strategy [video]

https://www.youtube.com/watch?v=VZsm3Z2njAQ
1•keepamovin•31m ago•0 comments

They Built the 'Cursor for Hardware.' Now, Anthropic Wants In

https://www.wired.com/story/schematik-is-cursor-for-hardware-anthropic-wants-in-on-it/
1•CharlesW•32m ago•0 comments

My Linux Setup for Work and Life – NixOS, Niri, Helix [video]

https://www.youtube.com/watch?v=CeUOz_xtO-o
1•AnthOlei•32m ago•0 comments

Show HN: Kern – Agents that do the work and show it

https://github.com/oguzbilgic/kern-ai
1•obilgic•35m ago•0 comments

Sony implementing age verification for PlayStation users

https://twitter.com/CR1337/status/2046427329866694676
3•CR1337•38m ago•1 comments

The Ferrari of Espresso Machines Is Fueling a Hot Resale Market

https://www.nytimes.com/2026/04/20/dining/la-marzocco-espresso-machine.html
3•mitchbob•43m ago•1 comments

Voice to Instrument

1•starkiron•44m ago•0 comments

Wormhall

http://iladelf.org/wormhall/index.html
1•madprops•44m ago•0 comments

Claude Desktop Works with OpenCode Go

https://gist.github.com/avarayr/a9a35354aa6d7d8430ce0c27cd9aff3f
1•mikamika83•44m ago•0 comments

Mathematician Collapses All Functions to One Weird Formula [video]

https://www.youtube.com/watch?v=hwtqJaS42xk
2•darepublic•53m ago•0 comments

The SF Group Chat

https://twitter.com/daniel_dhawan/status/2041913527045386447
1•nowflux•56m ago•0 comments

It's not just one thing – it's another thing

https://techcrunch.com/2026/04/20/ai-writing-its-not-just-this-its-that-barrons/
1•davikr•56m ago•0 comments
Open in hackernews

KV Cache Compression 900000x Beyond TurboQuant and Per-Vector Shannon Limit

https://arxiv.org/abs/2604.15356
43•EGreg•1h ago

Comments

tomrod•1h ago
Extraordinary claims! I don't follow the argument though.
EGreg•1h ago
Author here. Since starting to teach AI at IENYC, I started publishing my papers recently on arXiv, and considering submitting them to a journal.

This is based on my original "PLT" paper: Probablistic Language Tries (https://news.ycombinator.com/item?id=47743585). A "Trie" is basically a tree of prefixes. While working on https://safebots.ai I became obsessed with caching generated artifacts as a means to do a lot of things: extremely cheap inference, near-optimal compression, modeling decision trees for strategies, and so on.

The PLT model was about compression in general. My main insight there was that the LLM's own weights actually contain an incredibly detailed probability distribution of "the next token" in any sequence, which can therefore be very useful to supercharge statistical compression. Sequences which occur frequently in the domain of the model receive short codes. The other insight is that if we allowed lossy compression, we could compress well below the Shannon information limit, and just have an "overflow" bag for surprising sequences.

When TurboQuant came out, I realized we can also go way below the Shannon limit in the same way, and take advantage of PLT. In fact, I'm working on publishing a paper that generalizes this to robotics (which needs to do cheap fast on-board inference "in the field"). I also believe this is how animals actually learn. In other words, over time they learn overall "sequences" of actions and then can check whether they are "good enough" to solve the problem, or whether to switch to a full analysis -- this corresponds to System 1 and 2 of Daniel Kahneman's "Thinking Fast and Slow".

If you want more specific information, or see the code for a working prototype, you can write me at the email in the paper.

Rekindle8090•59m ago
I see you using a double dash instead of an em or en dash to get around bot detection extensions and I'm not fooled.
EGreg•58m ago
Haha, yes I always used -- when I typed an em-dash manually. What bot detection extensions? :-P
cristoperb•51m ago
I can't speak for the person you're replying too, but I use -- for emdash for two reasons: I never remember how to type an actual emdash in linux/X11, and more importantly, I do most of my writing in Asciidoc which converts -- to an emdash automatically. It's nothing to do with bot detection or whatever.

But it does get me confused sometimes because in LaTeX (and other markup languages) -- gets converted to an endash whereas it takes three hyphens --- to make an emdash.

rhet0rica•38m ago
you are hereby sentenced by the council of dashers to type "—" ten million times using Windows-1252 alt codes

you have 5 seconds to comply before your planet will be demolished to make room for a giant space-typographer's punctuation case

stingraycharles•52m ago
Dropping a grand theory of animal cognition into a defense of a KV cache compression bound is not something I was anticipating. I don’t think it’s a great argument.
wholinator2•41m ago
At least some random pseudocrackpotery like that is points in the direction of it being a human. There's some strange human tendencies that AI just doesn't usually replicate
usernametaken29•47m ago
Kahnemans book is considered outdated by modern neuroscience.
himata4113•40m ago
The reasoning around the 900000x claim isn't sound and violates way too many information density principles.

I was incredibly curious since I had a pet theory in my mind about something extremely similar, but arrived at a conclusion that the time complexity of such cache would end up being extremely slow.

This is like saying that you've achieved single token compression when you're passing a single token into a model and letting it regenerate the entire output since at the end of the day models are probabilistic stateless devices. At that point you don't have a cache and are just replaying the tokens or have a caching algorithm with a complexity similar to that of a model defeating the purpose of such cache.

I've never considered that arXiv had a problem, now I do.

EGreg•32m ago
No, the 914,000x in the paper is talking about the ratio between two entropy floors, it's not a claim about practical compression. The point is that per-vector quantization has been chasing the wrong theoretical limit: the sequential entropy bound is just fundamentally lower, by that factor, because KV vectors aren't independent samples!

On complexity, that's fair concern, and the paper doesn't fully resolve it. But the analogy to "replaying tokens through the model" isn't exactly right. The delta coding layer uses the model's own next-token prediction, which is already happening during normal autoregressive inference. You're not adding a forward pass, you're using the one already running and storing only the residual, which is much smaller than the raw vector -- precisely because the model is a good predictor of its own next state.

The trie index lookup is O(sequence length), not O(model forward pass). Whether that's fast enough in practice at scale is actually a legitimate open question and I'd be the first to admit the paper doesn't settle it. But the contribution here is simply establishing that the bound exists and is dramatically lower than what the field has been targeting. That's what I wanted to put out. The engineering question of how close you can get is the natural next step.

Your pet theory about time complexity sounds interesting actually, did you write it up anywhere?

mbernstein•34m ago
This is a compute memory trade, not compression vs. turobquant? Lemma 1 is something like, "forward pass is deterministic because it's deterministic" which means the input tokens were always the lower bound...which isn't caching? Smells tautological. What am I missing?
EGreg•26m ago
Well yeah, I just wrote it as a lemma, but it's basically close to tautological. Its only job is to formally ground the entropy argument that follows it. The interesting claim is what comes after: because KV vectors are deterministic functions of tokens, and because the model is a near-optimal predictor of its own distribution, the conditional entropy of each new KV vector given all previous ones is bounded by token-level perplexity. TurboQuant compresses against the marginal distribution of each vector in isolation -- that's the gap.

And yes, it's a compute/memory tradeoff, all caching is. The claim is just that the memory floor is much lower than anyone had formally established. Whether the compute cost of getting there is worth it is a fair open question the paper doesn't settle. But what if it is? Caching is the thread running through most of my work, and I intend to find out.

gaze•56m ago
this paper looks AI generated to me... I mean, there's no experiments to go along with it.
ddtaylor•1h ago
Very intersting. A compression strategy that uses the model itself as the dictionary.
thethirdone•58m ago
> the ratio remains approximately 914x over TurboQuant, with compression improving rather than degrading as context length grows.

This line from the abstract got me really suspicious. Obviously a compression scheme that incorporates the entire sequence shouldn't get worse compared to a per element one as the length increases.

It is important to note that this paper is PURELY theoretical. I couldn't find much meat on the bone from a quick skim.

The single author, Gregory Magarshak, has only published one paper on arxiv before and appears to be a professor of business / music. I don't plan to give it more of a read hoping for something of value.

stingraycharles•50m ago
Me neither. There are no actual experiments / data, no peer reviews, and the innovation relies almost entirely on citations from the author’s other paper.

The author is not an ML researcher but rather an AI startup CTO / founder. Previously worked on “social operating systems” for the web, blockchain of course. And now an AI innovator. I’m suspicious. This was part of the author’s reply in another thread:

> When TurboQuant came out, I realized we can also go way below the Shannon limit in the same way, and take advantage of PLT. In fact, I'm working on publishing a paper that generalizes this to robotics (which needs to do cheap fast on-board inference "in the field"). I also believe this is how animals actually learn. In other words, over time they learn overall "sequences" of actions and then can check whether they are "good enough" to solve the problem, or whether to switch to a full analysis -- this corresponds to System 1 and 2 of Daniel Kahneman's "Thinking Fast and Slow".

Which doesn’t exactly inspire confidence and makes me wonder who they think their audience is. ML researchers or LinkedIn.

gaze•49m ago
the irritating thing about LLM generated papers like these is that they're wrong but are generated using LLMs that are capable enough to bury the absurd claim pretty deep in there.
stingraycharles•47m ago
Analyze it using an LLM. Claude was pretty ruthless about this one.
gaze•43m ago
sure but it seems spiritually wrong to use an LLM to debug a slop paper. Who knows, maybe claude generated it in the first place?
thethirdone•41m ago
Yeah, for me Claude identified the phrase "this holds with probability 1 over random weight matrices since the null space has dimension"

Treating trained weights as random for the purpose of a proof is immediately discrediting for a paper to me.

EGreg•17m ago
"This holds for almost all matrices" is actually something you'd want to know if we're talking about probabilities, no?
EGreg•38m ago
You're right, I'm not a well-known researcher, simply an entrepreneur who started to publish academic papers.

However, I do have a long history of diving deep into fields and building practical, open-source solutions to major problems I perceive in the fields.

15 years ago I started with social networks and PHP: https://github.com/Qbix http://laweekly.com/restoring-healthy-communities/

8 years ago I got into smart contracts on EVM, which was the SOTA at the time: https://github.com/Intercoin https://intercoin.org/applications

About a year and a half ago I started teaching a course on AI at a university not far from NYU where I studied... and that's what got me into this: https://vimeo.com/1063008765/c7ef3abcc5

I try to document everything on GitHub and popular articles, but only recently started publishing academic papers on arXiv and plan to actually start submitting them for real publications. While I build, I realized that I should start publishing any novel theoretical results that underpin my work.

I plan to publish actual code in a few weeks. To be fair, TurboQuant is also a purely theoretical paper. I just wanted to get this out and share.

thethirdone•29m ago
> To be fair, TurboQuant is also a purely theoretical paper. I just wanted to get this out and share.

TurboQuant is not a purely theoretical paper. Section 4 "Experiments" (page 15) [0] has a bunch of figure based on actual GPU computations.

[0]: https://arxiv.org/abs/2504.19874

stingraycharles•15m ago
TurboQuant went through ICLR review, has multiple Google Research co-authors, open-source implementations, CUDA kernels, and LongBench benchmarks.

Contrast that with your paper: no experiments, no implementation, no empirical validation of any kind.

Did you try engaging with LLM researchers and get their feedback on your paper?

sabareesh•53m ago
Sounds like speculative decoding but for KV cache
aesthesia•46m ago
> The second layer, predictive delta coding, stores only the residual of each new KV vector from the model's own prediction of it

I don't understand this. The key and value vectors for any given layer + token are created by the model. By definition, they are exactly equal to the model's prediction of them!

Extreme KV cache compression is easy to get---you can get an infinite compression ratio by just regenerating the key and value vectors on every forward pass. The point of a KV cache is to reduce the amount of repeated computation during generation, though. Compression only helps if you have an efficient decompression algorithm.

binsquare•42m ago
Incredulous claims and unreviewed paper.

Attention really is all you need.

CyberDildonics•37m ago
I think you mean incredible claims. You would be incredulous about them.
EGreg•28m ago
The prediction being used is the model's prediction of the next token's KV vector, given all previous KV vectors. Because the model was trained on language, it has strong priors about what comes next. The residual, i.e the difference between the predicted next KV vector and the actual one -- is much smaller in entropy than the raw vector, for the same reason language model perplexity is low on fluent text.
aesthesia•15m ago
What model is doing this prediction? The only way a transformer predicts the "next KV vector" is by sampling the next token and then running a forward pass with that token.