nmca

6 ms·

31.
by nmca 1y ago
hardware tokens are the way! Everyone has had a house key their whole lives, and understands how to keep a spare to prevent lock-outs.
32.
by nmca 1y ago
This would be much better with an example
33.
by nmca 1y ago
Only tangentially related really but almost all of the variance I have ever observed in ability to remember things like this is just trying at all to use a mnemonic or routine. The above routine is a specific practice to remember somethin
34.
by nmca 1y ago
The original dream of crypto was a good one.
35.
by nmca 1y ago
The linked USAMO math results are in an exam that requires proofs. The same authors, on the same website, ran AIME 2025 shortly after it happened and found it was totally consistent with the o1 announcement numbers; the difference being tha
36.
by nmca 1y ago
This whole thread is true consistent statements that differ only in emphasis.
37.
by nmca 1y ago
in 2019 these things could hardly speak, in 2024 we trained one that outperforms all but the very best in the world at competition programming. the bull case very obviously speaks for itself!
38.
by nmca 1y ago
A truly wonderful novel that I also opened the comments to shill.
39.
by nmca 1y ago
Been using elevenlab reader, but these are much better!
40.
by nmca 1y ago
very big if true! I downloaded their data from the metr public evals GitHub repo and independently re-implemented the very basic maximum likelihood analysis, which gave very similar results to what they shared.
41.
by nmca 2y ago
The authors own summary from the position paper is: In particular, we examine the one-step iterative process of DDIM [39, 19, 21] and show that it has limited capacity with respect to the target timestep under the current denoising network
42.
Write a metafictional story about AI and grief(twitter.com)
3 points by nmca 2y ago|2 comments
43.
Pgserver: Super easy pip-installable Postgres(github.com)
2 points by nmca 2y ago|0 comments
44.
by nmca 2y ago
As an aside, an FM index can be used to efficiently turn an LLM into an actual stochastic parrot (one that emits only substrings of some dataset). This is more useful than it sounds because you can use it for quoting from large corpora.
45.
by nmca 2y ago
Whilst brexit was a bad idea —- look forward not backward! Planning reform is a viable opportunity and making it as good as possible is a much better use of energy than moaning about mistakes past.
46.
by nmca 2y ago
I’ve had three show-stopping bugs across core apple software in the last month — I wish the competition was better!
47.
by nmca 2y ago
This is grok 3, so not a debut
48.
by nmca 2y ago
I’ve listened to a few audiobooks on long drives, and have been surprised how hard it is to find good voices on audible. Often a book that might otherwise be good has a prohibitively annoying tone. So honestly the exciting thing here is the
49.
by nmca 2y ago
The discourse around synthetic data is like the discourse around trading strategies — almost anyone who really understands the current state of the art is massively incentivised not to explain it to you. This makes for piss-poor public epis
50.
by nmca 2y ago
(while I work at OAI, the opinion below is strictly my own) I feel like the current version is fairly hazardous to students and might leave them worse off. If I offer help to nontechnical friends, I focus on: - look at rate of change, not c
51.
by nmca 2y ago
It’s about 1oom off. In some contexts, one oom is pretty close.
52.
by nmca 2y ago
I for one am hugely supportive of compute that is red white and blue.
53.
by nmca 2y ago
Can you define what you mean by novel here?
54.
by nmca 2y ago
The real turker studies, resulting in the ~70% number, are scored correctly I believe. Higher numbers are just speculated human performance as far as I’m aware.
55.
by nmca 2y ago
75% of 400 is 300 :)
56.
by nmca 2y ago
It is allowed exactly two guesses, per the ARC rules.
57.
by nmca 2y ago
I can’t provide more information than is currently public, but from the ARC post you’ll note that we trained on about 75% of the train set (which contains 400 examples total); which is within the ARC rules, and evaluated on the semiprivate
58.
by nmca 2y ago
We worked with ARC to run inference on the semi-private tasks last week, after o3 was trained, using an inference only API that was sent the prompts but not the answers & did no durable logging.
59.
by nmca 2y ago
You’re trying to to characterise personal risk and they were tying to characterise societal risk; both are valid.
60.
by nmca 2y ago
The problems with python types are: A) The type system is fairly bad, as type systems go (forgivable) B) the type checkers are, for large codebases, excruciatingly slow, to the point that the tests are faster!! The second is not forgivable.

More ›