benob

6 ms·

31.
by benob 7mo ago
What about doing that with movies and music?
32.
by benob 7mo ago
Also, don't forget to install our local ad blocking LLM. Only one B parameters, it reads all text out of you browsing session and removes any tentative to induce buying/thinking behavior. Best in town...
33.
by benob 7mo ago
I had that in mind too. What if you handcraft a subnetwork with (some subset of) Turing machine capability? Do those kinds of circuits emerge naturally during training? Can reasoning use them for complex computation?
34.
by benob 7mo ago
What is the need for dependencies when you can code them from scratch?
35.
by benob 8mo ago
Same with asbestos, I mean what could go wrong in 20 years?
36.
by benob 8mo ago
You could totally make a believable timing generation model from a few (hundreds) recordings of human writing. Detecting AI is hard...
37.
by benob 8mo ago
Give me self hosting: LLM generates compiler which compiles LLM training and inference suite, which then generates compiler which...
38.
by benob 8mo ago
Why don't LLMs directly generate machine code?
39.
by benob 8mo ago
I really like the "file lens" example: > “Focus on…” would allow the user to specify what they're interested in changing and present only files and lines of code related to their specified interest. > “Edit as…” would a
40.
by benob 8mo ago
What local models shine as local assistants? Is there an effort to evaluate the compromise between compute/memory and local models that can support this use case? What kind of hardware do you need to not feel like playing with a useles
41.
by benob 8mo ago
Too bad it doesn't have zooms echo cancelation / isolation
42.
by benob 9mo ago
Maybe it will also change the whole publication as evaluation of science.
43.
by benob 9mo ago
An LLM targeting language and no token efficiency?
44.
by benob 9mo ago
It was over when they named the company Open AI
45.
by benob 10mo ago
Is there a pip installable version?
46.
by benob 10mo ago
What, no reference to quantum or crypto?
47.
by benob 10mo ago
Will there ever be a llama12? Is it going to go the yolo route?
48.
by benob 11mo ago
How was the attack detected in the first place?
49.
by benob 11mo ago
Was it this one? https://infini-gram.io/
50.
by benob 11mo ago
What does it mean nowadays to start from scratch? At least in the open scene, most of the post-training data is generated by other LLMs.
51.
by benob 11mo ago
If caching is bounded in time, can't you use other fingerprinting methods to seal the gaps?
52.
by benob 11mo ago
Why doesn't this apply to any kind of cached content?
53.
by benob 11mo ago
Google's move is very good for the web. By pushing app makers away from walled platforms, you turn them to standardized, open ones such as the web.
54.
by benob 11mo ago
> Bring Your Own Language Few-shot new languages is going to be a game changer for linguists
55.
by benob 11mo ago
Not tested on that particular model, but the idea has been flying around for some time: https://arxiv.org/abs/2509.04166v1
56.
by benob 11mo ago
It comes with a basic interpreter, but the thriving community develops plenty of stuff: python, lua, forth, lisp... have nice ports which you can play with on device. There also is a library of software developed for rp2040 which have been
57.
by benob 11mo ago
I have been having a lot of fun with PicoCalc. It's not targeted at end users but is fun for developers alike who want a taste of developing things from first principles. More than anything it can live independently from your other dev
58.
by benob 11mo ago
This is very lenient French: "fetchez le dico"
59.
by benob 1y ago
Audio tokenization consumes at least 4x tokens versus text. So there is an efficiency problem to start with. Then is there enough audio data to train a LLM from scratch?
60.
by benob 1y ago
Is there anything preventing them from using heterogeneous memory chips, like 1/4 GDDR7 and 3/4 LPDDR? It could enable new MEO-like architectures with finer-grained performance tuning for long contexts.

More ›