olq_plo

Dunno why people still write about Kokoro, there are better alternatives. Pocket TTS has more natural intonation and voice cloning. Supertonic 3 is another great model in the same weight class, which is the only one that can autodetect language and make a mix of different languages sound good.

Dumb. This is what modules are for. Also, the stdlib is extremely well designed. It considers edge cases most people never think about. Source: I am a Boost Developer.

That blog post is full of AI slop. Repeats the same argument a gazillion times. It's not X, it's Y. Awful to read.

Great, thanks for making me Google what CRM means in this context. Neither your post nor your website explains the acronym.

And now you can do this with polars in parallel on all your cores and the GPU, using almost the same syntax as in pyspark.

Corollary: Doing the thing and not talking about it in a hammer tweet is also 'not doing the thing'.

I wrote my first post about parsing webpages to structured data with a LLM in January, using local models. Now its October and I did it again with current models and libraries. Boy, what a difference.

The post is so painfully obviously AI written, it hurts my eyes.

The Setup

The Scoop

The Conclusion

I hate AI slop.

Yes, bad article to omit that. It is such a cool fun fact. Gauss was unreal.

Since you seem to know your stuff, why do LLMs need so much data anyway? Humans don't. Why can't we make models aware of their own uncertainty, e.g. feeding the variance of the next token distribution back into the model, as a foundation to guide their own learning. Maybe with that kind of signal, LLMs could develop 'curiosity' and 'rigorousness' and seek out the data that best refines them themselves. Let the AI make and test its own hypotheses, using formal mathematical systems, during training.

IANAL, but it means the commit itself is public domain. When integrated into a code base with a more restrictive license, you can still use that isolated snippet in whatever way you want.

More interesting question is whether one could remove the GPL restrictions on public code by telling AI to rewrite the code from scratch, providing only the behavior of the code.

This could be accomplished by making AI generate a comprehensive test suite first, and then let it write the code of the app seeing only the test suite.

Very cool idea. Can't wait for converted models on HF.

This is not everyone, but if you've been writing scientific papers with LaTeX, you may have come across this issue.

You go to an online database (Inspire or ADS) to fetch some references for your paper. Then you have to copy/paste the entry twice, the key to your LaTeX document and the BibTeX entry to your .bib file. Doing redundant things is annoying, right? autobib removes the need to do the latter. You still have to look up the key online and cite it in your LaTeX document, but autobib downloads the entry automatically to your .bib file.