1. benob Dec 16, 2025 parent

   Is there a pip installable version?
2. benob Dec 9, 2025 parent

   What, no reference to quantum or crypto?
3. benob Dec 9, 2025 parent

   Will there ever be a llama12? Is it going to go the yolo route?
4. benob Nov 24, 2025 parent

   How was the attack detected in the first place?
5. benob Nov 20, 2025 parent

   Was it this one? https://infini-gram.io/
6. benob Nov 18, 2025 parent

   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.
7. benob Nov 17, 2025 parent

   If caching is bounded in time, can't you use other fingerprinting methods to seal the gaps?
8. benob Nov 17, 2025 parent

   Why doesn't this apply to any kind of cached content?
9. benob Nov 13, 2025 parent

   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.
10. benob Nov 11, 2025 parent

    > Bring Your Own Language

    Few-shot new languages is going to be a game changer for linguists

11. benob Nov 11, 2025 parent

    Not tested on that particular model, but the idea has been flying around for some time: https://arxiv.org/abs/2509.04166v1
12. benob Nov 5, 2025 parent

    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 ported, such as a mac classic emulator. If you feel that a micro controller is too low power, you can plug in a luckfox lyra which runs a proper linux with 128M of RAM.
13. benob Nov 5, 2025 parent

    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 devices.
14. benob Oct 31, 2025 parent

    This is very lenient French: "fetchez le dico"
15. benob Oct 21, 2025 parent

    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?
16. benob Oct 15, 2025 parent

    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.
17. benob Oct 10, 2025 parent

    This work is a good argument against memorization of information seen less than 250 times during training.
18. benob Sep 26, 2025 parent

    I wonder if one could store only the binary representation at training and sample a floating point representation (both weights and gradient) during backprop.
19. benob Sep 23, 2025 parent

    There was a time when people would estimate n-gram probabilities with feed-forward neural networks [1,2]. We just improved that with the (multilayer) attention mechanism which allows for better factoring over individual tokens. It also allowed for much larger n.

    [1] https://jmlr.org/papers/volume3/bengio03a/bengio03a.pdf

    [2] https://www.sciencedirect.com/science/article/abs/pii/S08852...

20. benob Sep 23, 2025 parent

    Like it or not, LLMs are effectively high-order Markov chains
21. benob Sep 23, 2025 parent

    These techniques are limited to structures that can be checked with bounded history or bounded memory (that can be checked with a grammar or FSA). What about more complex structures that don't factor easily?
22. benob Sep 20, 2025 parent

    My understanding was that the Alpaca data was a distillation from text-davinci-003
23. benob Sep 18, 2025 parent

    a good benchmark for video understanding in IA
24. benob Sep 16, 2025 parent

    It's funny because automation is the only thing you can expect from AI
25. benob Sep 9, 2025 parent

    > YouTube views seem to have fallen off a cliff recently

    So they started discounting AI data collection bots?

26. benob Sep 4, 2025 parent

    Stop thinking about text being the data. There are so many other sources, even some that you can generate.

    https://arxiv.org/pdf/2506.20057

27. benob Aug 27, 2025 parent

    Even if it's not on topic, that post is quite interesting.
28. benob Aug 1, 2025 parent

    How is it different from mere syntactic sugar over the same programming concepts? What does it bring that C++ cannot do?

    Isn't it just a way of controlling the language vs using normative bodies?

29. benob Jul 23, 2025 parent

    Are there efforts to include the neccessary context in compilers to autovectorize?
30. benob Jun 19, 2025 parent

    I wonder how difficult it would be to bias a model so that it subtly corrupts election results when performing OCR.

This user hasn’t submitted anything.