178 points by gmays 2 days ago | 76 comments | View on ycombinator
t0mek 2 days ago |
eddiewithzato 2 days ago |
svnt 2 days ago |
In January it seemed very clear the subscription models behind claude code became less competent, there was a whole series of posts about it. I stopped working on it as a result.
HackerThemAll 2 days ago |
I am certain that all these companies have stored and backed up all they could find, so that it can't be taken away from them.
octoberfranklin 2 days ago |
zodwick 2 days ago |
consumer451 2 days ago |
highfrequency 2 days ago |
mlmonkey 1 day ago |
troupo 2 days ago |
kazinator 2 days ago |
undefined 2 days ago |
LarsDu88 1 day ago |
suddenlybananas 2 days ago |
pcrh 2 days ago |
When would one want to predict a full-length unknown protein sequence from only a fragment? Any fragment of any useful length can be back-translated to the DNA sequence and that can be tested against genomic databases (a process called tblastn). And if your protein/gene is not found there, you have no way of knowing whether or not the predicted sequence has any basis in reality.
Testing the model against shuffled sequences is also kinda cheating, as these are no longer random pool of amino acids, but one whose source already exists in the databases.
spwa4 2 days ago |
And electrostatic interactions.
And some methylation.
And some phosphorisation (but now we're getting deep into somewhat rare grammar)
Yeah, DNA is a language that is way too complicated for human minds to understand, although plenty of humans "get the gist" so to speak after looking at it for a few hours, but ... here's some bad news: it's only somewhat too complicated for LLMs. So why wouldn't you train LLMs on them? Frankly, it's the one language we have "texts" from that is actually difficult to understand for LLMs and ... well, it's definitely improving their understanding.
(also we have quite a bit of DNA text that isn't from the internet or books, it's freely downloadable and unlike everything else LLM companies steal, actually not covered by copyright)
Filipc 2 days ago |
The shuffled-sequence control is particularly compelling. It suggests that the models aren’t simply recalling amino acid frequencies or reproducing patterns from scientific literature. They’re capturing structural information embedded in the sequences themselves.
There’s an important distinction, though, between learning statistical representations of protein families and actually understanding the underlying biology. Next-amino-acid prediction is a fascinating proxy, but it doesn’t necessarily demonstrate that the model has learned the mechanisms governing protein folding or function.
I’d be particularly interested in seeing whether these capabilities transfer to genuinely novel protein families, rather than previously unseen members of known families. The post-cutoff experiment is a good step in that direction, although it still tests generalization within existing families.
Either way, this raises an interesting question about foundation models: how much latent scientific knowledge are we inadvertently creating simply by exposing them to sufficiently large and diverse datasets?
Perhaps we’re underestimating what can emerge from learning the statistical structure of the natural world.
gigiamoroso 2 days ago |
Aside from the obvious applications in biochemistry, could protein sequences also be used as a new training data source to improve overall model quality? If that's true, it's quite fascinating that we can take highly complex data we don't fully understand and use it to improve LLM training, simply because it contains so many significant structures and relationships.