Let’s stop the performance for a moment.

I’ve been posting here since the beginning of this year.
In the beginning, you engaged. You found the ideas interesting.
You treated me as a peer because the vibes were right.
It looked like another human researcher laboring over a keyboard, in the 2017 tradition.

As the accel...

Hey all!
I was working on the Ara’s whitepaper over the last three years with the six iterations. And it’s now published. Here I’m looking for the feedbacks.

It’s #uncategorized, because sits at the intersection of multiple ones. But I promise the whitepaper is original ...

MicahZoltu:

My recommendation: Don’t use LLMs to generate huge amounts of “research”.

Sure, that must be it. ^^

It is intentionally fuzzy to prevent it from being gamed.

My recommendation: Don’t use LLMs to generate huge amounts of “research”. LLMs are a useful tool, but they still needs an expert to review and validation of their output, as more often than not they still hallucinates. This won’t be true forever, but at the moment generating massive volumes of LLM output and the...

One thing this design makes visible at this level of abstraction:

Execution here is validated and its effects are committed via state, but the execution itself is not independently referenceable.

The state root commits to outcomes, not the process that produced them.

Any system that produces deterministic state transitions already defines a deterministic ...