cheesecompiler

depression, like alcoholism, is a crutch and a symptom of a deeper issue, not an independent behaviour in a vacuum.

The authors take a leap in assuming there's a difficulty with overriding a response. But in my and others' experience from group therapy, I can confidently say it's more like either apathy or being numbed out, to the point of either not noticing or caring that the aversive sound is happening.

This is how SSRI has felt for me also, incidentally: an even deeper apathy, setting in from lack of any emotion at all.

The walls of the Google garden have been slowly going up.

Wow no one at all with a Vision Pro?

In my experience Gemini is consistently more conservative and poor at reasoning/regurgitates, like a local Llama instance.

But this uses the transformers model to justify its own reasoning strength which might be a blindspot, which is my original point. All the above shows is that transformers can simulate solving a certain set of problems. It doesn't show that they are the best tool for the job.

The overarching point still stands: our formal systems are just models built to describe the patterns we observe. In that sense, math “just is.” The fact that some models aren’t compatible with others doesn’t undermine that—it just shows they’re incomplete or context-dependent views into a larger structure.

The reverse is possible too: throwing massive compute at a problem can mask the existence of a simpler, more general solution. General-purpose methods tend to win out over time—but how can we be sure they’re truly the most general if we commit so hard to one paradigm (e.g. LLMs) that we stop exploring the underlying structure?