adw

Wildly, severely not, honestly. They're going about retail second-hand on Reverb, with thin supply. You might do better having a look in the Chase Bliss Audio Discord.

It's not a ROMpler, but Hydrasynth Explorer.

Melanie Matchett Wood is elliptic curves, which Anthropic got some results on recently.

I’d be shocked if anyone with any kind of post-secondary education in any numerate discipline couldn’t give you at least an informal definition of a metric space. Certainly all the physicists, all the geneticists, all the ML people.

Fidji Simo has well documented chronic health issues so there’s no reason to assume that the ostensible and actual reasoning are far apart there.

The famous levitating frog guy has the double; one Nobel, one IgNobel.

Schools use it!

The Noah Smith/Nate Silver axis of opinion-politics writers are — literally — Stephen J Smith and Skip Bayless for people who think politics is sports. They're in the entertainment business.

There are many examples; k12.<state>.us is another.

The implication (given that the US is an outlier here) is that prevention is much better than cure here; that means reform of healthcare and social housing so people are less likely to wind up in that trap.

> Yet IPv6 hasn’t been adopted.

news to literally anyone on a mobile internet connection (which is most internet connections!)

What is the error rate of human programmers? Anyone who tells you that either human or A.I. code is magically exempt from issues is selling you something.

Other post-train mechanisms have meaningful bandwidth for introducing new information to the model (on-policy distillation is more or less the other extreme). RLVR aligns the model to particular behaviors it could already (unreliably, intermittently) express, it doesn't introduce new behaviors; the behaviors were latent in the pretrain/midtrain.

This paper is less dramatic than you think it is and really just re-explains what RLVR does.

Let's stipulate that what pretraining does is train next token prediction over a gigantic corpus. You can then sample from this distribution repeatedly (cf the Large Language Monkeys paper) and count how often it passes some deterministic verifier.

What GRPO-style RLVR does is precisely this, but then reward the trajectories which passed the verifier. These distributions are _by construction_ within the accessible output space of the pretrained model; you're reweighting the distribution so that pass@k goes up, because that's (for applications like programming) very useful. RLVR is about making sampling more efficient; the only new information being added to the system is the presence of the verifier, and note that you only get a reward when the verifier passes, so there's essentially no mechanism for "teaching new facts" here.

She went to UCL which is also a tier one university, though.

One of the coolest things about seeing a show at the Fillmore is getting the posters they hand out. A number of bands sell, or have their artists, sell show posters (which then become collectible limited editions); it's way less offensive when it's King Gizzard and the Lizard Wizard than when it's Wizards of the Coast.

I mean, they kind of have, but mostly by finding an infinite money glitch.

They’re skint too. It’s something you would really like the research council to take on.

The living-costs stipend for an EPSRC PhD student is around £20k so it's about ten percent of that... big commitment for a student to make!

The usual answer is "config macros", because if you don't give people something Turing complete they'll just m4 their way into Turing completeness and that isn't really an improvement. (Starlark in Bazel, etc.)

Another intuition is that near a minimum you can Taylor expand the function and show that the higher order coefficients (past the square) are negligible.

It is common for big workplaces to have multiple unions and essentially all unions are sectoral and role-specific rather than company specific.

Take the NHS; it will have to deal with ten plus separate unions - https://nhsunions.org/#about – of which the biggest powers are the British Medical Association and the Royal College of Nursing, but the cleaners are GMB or Unite and they're huge pan-working-class institutions.

Education has to deal with the NEU, the NASUWT, and the NAHT, each of which has a different political slant. Some unions in the UK have been historically rather centrist in their politics (a good example of that is Prospect, https://prospect.org.uk/about/, which is a roll-up of a number of scientific and finance unions), some are firebreathing communists, but all of them work across employers.

There's also no such thing as a closed shop in the UK – because there are much stronger worker protections, there's less of a need for one.

(I was, at one time, in a majority-UCU workplace; https://www.ucu.org.uk.)

> you cannot go to a bank and ask them to help you acquire something 100x worth your cap.

you literally can, it's the entire point of eg management buyouts

It's just an acronym, "parameter-efficient fine tuning". LoRA is one method, prefix tuning is another, there are more.

> So much precision is required that session musicians are playing most of the things you hear, not the actual artists.

In pop music this has been true since the 60s. For independent music it has mostly never been true. This hasn't changed much.

Leverage. They’re essentially an 80s style junk bond LBO house.

kexp.org in Seattle and San Francisco for a start.

https://podcasts.apple.com/us/podcast/machine-learning-stree... is a pretty good primer on METR, what it measures, and its limitations.

arXiV is not intended to be your blog. You should be held to a zero-mistake standard when publishing academic work.

The people I worry for are the junior researchers who are going to be splash damage for dishonest PIs. The PIs, though, deserve everything that’s coming for them.

You are forgetting the albums KG and LW which are both microtonal. I knew exactly what you were saying :-)