proof_by_vibes

Needed this. Thanks.

Let's see what the common denominator has to say about this!

I really think this gets at the heart of the distinction between language as it pertains to how we connect with others and language that records our observations of the world and its history.

I think the ideal mode for engaging with LLMs should be as interactive encyclopedias. They are excellent at rewarding curiosity when someone passionate about learning sits in the driver seat of this kind of tool. There is something to be said about the benefits of _active_ learning over _passive_ instruction.

Nonetheless, it's impossible to consistently discern fact from fiction when in the form of narrative. As such hallucinations are the key counterexample for why we can't have a world where LLMs can be effective instructors. Because it's one thing to be exposed to the structure of a narrative, and to read it as grammar, but you don't ever know how to _feel_ about these structures until you become tuned to another human being.

This is a very male attitude to have.

Perfectly safe. I would argue that it is the safest of the three, the least invasive both in terms of its design and in terms of privacy.

The open source model of development has encouraged the correct incentives for people to become active in identifying and fixing possible exploits in a global, communal effort.

Every server on the cloud has (by a large margin) chosen Linux as the OS to trust specifically for this reason.

I would not go as far to say that they exist in a bubble. I quit my job as an engineer because this exact sentiment from my boss was ruining my life.

I'll take three hours instead and hop on a train to get brunch.

I agree that it is relative, but disagree with your conclusion. I think the relativity you have in mind is what we normally think of as a setting.

I've been iterating on sodium bindings in Lean4 for about four months, and now that I've gotten to Ristretto255 I can see why the author is excited about its potential. Ristretto is a tightly designed API that allows me to build arbitrary polynomials on Curve25519 and I've been having a blast tinkering and experimenting with it! If the author by chance reads this, just want to say thank you for your work!

This is perfect. I'm currently creating a MUD and these are exactly the kind of fonts I want. Thanks for sharing!

Are there any experts that could help me bootstrap myself on the current literature on "world models?"

I'm of the opinion that formalization is the biggest bottleneck of current generation LLMs. However, I don't think that this necessarily suggests that LLMs don't benefit from formal methods. Given existing abstractions, Lean4's exceptional tooling allows for more efficient iteration with LLMs and requires less human supervision since Lean's language server provides specific and actionable feedback whenever the LLM makes a mistake. I've also noticed that LLMs can reason about code written in Lean4 far more effectively than in Python, despite Python having orders of magnitude more training data than Lean.

Nonetheless, I concur that LLMs don't yet know how to translate a request stated in a prompt to a complete Lean4 interpretation. My practice so far has usually required me to first choose an existing reference file that is similar to my desired goals, and use this reference as "inspiration" for how the LLM should go about formalization.

Finally I find this argument. Agreed, and I'm baffled that people think that AI is what's going to "solve loneliness." Loneliness has already been solved by YouTube/Twitch. The brain is easily tricked into thinking that it is "being social" when it is subject to the effects of the parasocial relationships that are formed by these platforms. People's afternoons are rapidly becoming consumed by hours of YouTube where they come out of it with a brain telling them: "boy, that's enough social interaction for today!" Introversion has become an epidemic as a result.