ac1spkrbox

The website is often user-hostile, in hopes of pushing you to the app.

Multimodal models are useful for lots of things! They can accomplish a range a tasks from zero-shot image classification to helping perform Retrieval-Augmented Generation on images. Like many generative model, I find the utility comes not necessarily from outperforming a human, but from scaling a task that a human wouldn't want to do (or won't do cheaply).

The set of techniques for retrieval is immature, but it's important to note that just relying on model context or few-shot prompting has many drawbacks. Perhaps the most important is that retrieval as a task should not rely on generative outputs.

“Any man who must say ‘I am the king’ is no true King”

I’m much more interested in my actual work than personal projects. Is that so weird?

And it should go without saying that sharing employer code could be considered theft.

I'm confused by this comment.

To me, Garmin is the only company here that does a good job with the "hardcore fitness" market, but why does that matter in the first place? The better market to compete in is the "fitness amateur" market, which I see as being many times the former in size. Fitbit competes well here, and Google has no offering to speak of. Seems like a successful diversification move for Alphabet's portfolio. I especially like what Google's software can do to improve FitBit's offering by leveraging AI.