Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Truth is not a direction: a Tarski attack on LLM probes

abeljansma.nl

Post reply on HN
Print view View original

81–88 of 88 posts

Re: Truth is not a direction: a Tarski attack on LLM probes

#81

Post by fourgreen »

TZubiri wrote: post #26
aesthesia wrote: post #22

Earlier quoted context omitted.

Quines produce similar issues to self-referential sentences without being directly self-referential. e.g. "'Yields falsehood when preceded by its quotation' yields falsehood when preceded by its quotation."

> "'Yields falsehood when preceded by its quotation' yields falsehood when preceded by its quotation." Here be the limits of my brain. It does feel (vibes) like an obfuscation of the self-referential canonical 'This sentence is false' example, like a sum of obfuscation techniques that are designed to confuse, but don't materially change the nature of the phenomenon: 1- reference is made implicit, instead of explicit…

So you say that: 'Yields falsehood when preceded by its quotation' yields falsehood when preceded by its quotation. does not have any meaning? Why? I think that both "'Yields falsehood when preceded by its quotation'" and "yields falsehood when preceded by its quotation" have very clear meaning, and the whole sentence also does. The sentence very clearly and precisely mean something, just it is neither true or false.

Re: Truth is not a direction: a Tarski attack on LLM probes

#82

Post by pocksuppet »

smallnix wrote: post #38
kstenerud wrote: post #21

> It might seem absurd to you to even suggest superhuman AIs could function as a truth-oracle (it certainly does to me), but there are two reasons to take it seriously. First, it is how these things will be used practically by the vast majority of people. They are already replacing standard Google search results, and I’ve had many discussions end with people delegating final authority on the truth to an AI. There are…

> The set-up assumes that the game and life are the same thing, and such is the pervasive nature of the idea of the game within the society that just by believing that, they make it so. (The player of Games, banks)

The Player of Games is a book about the stock market.

Re: Truth is not a direction: a Tarski attack on LLM probes

#83

Post by TZubiri »

fourgreen wrote: post #81
TZubiri wrote: post #26

Earlier quoted context omitted.

> "'Yields falsehood when preceded by its quotation' yields falsehood when preceded by its quotation." Here be the limits of my brain. It does feel (vibes) like an obfuscation of the self-referential canonical 'This sentence is false' example, like a sum of obfuscation techniques that are designed to confuse, but don't materially change the nature of the phenomenon: 1- reference is made implicit, instead of explicit…

So you say that: 'Yields falsehood when preceded by its quotation' yields falsehood when preceded by its quotation. does not have any meaning? Why? I think that both "'Yields falsehood when preceded by its quotation'" and "yields falsehood when preceded by its quotation" have very clear meaning, and the whole sentence also does. The sentence very clearly and precisely mean something, just it is neither true or false.

It might, I'm just saying it's beyond my comprehension. I might briefly grok it if I focus hard enough, but it will soon elude me again.

Out of curiosity, is the meaning different than "this sentence is false?"

Re: Truth is not a direction: a Tarski attack on LLM probes

#84

Post by Ukv »

HappMacDonald wrote: post #78
Ukv wrote: post #63

Earlier quoted context omitted.

ziofill's claim was that "A direction [in an LLM's embedding vector space] that is 99.99% accurate" is fine for practical purposes, not that 99.99% is fine for the chance of any given bite of food not killing you or similar hypotheticals - you'd want a few more 9s there. To justify relevance of inability to correctly answer liars-paradox-type questions ("what won't your response to this be?"), the article suggested t…

I think you are making a lot of assumptions on what "practical purposes" even mean in that case. Please think carefully and then try to tell me whether or not some Trump-administration government agency would not in the "99.99% reliable" circumstance just plug an LLM into the nukes and funnel worldstate input into it and have it make the decision "is it time to fire the nukes?" over and over again each second. I argu…

> plug an LLM into the nukes and funnel worldstate input into it and have it make the decision "is it time to fire the nukes?" over and over again each second [...] I argue that that would require far more nines than even "will food turn to poison in my mouth" would.

Sure - but (even assuming that's a practical purpose) the point is it that it doesn't need to be a 100% accurate truth oracle, which is all the article's argument prohibits. If the current human chain of command has 99.99999994% accuracy, then 99.99999995% accuracy is an improvement and not ruled out by the argument.

Re: Truth is not a direction: a Tarski attack on LLM probes

#85

Post by 83642736392 »

igleria wrote: post #37
kstenerud wrote: post #21

> It might seem absurd to you to even suggest superhuman AIs could function as a truth-oracle (it certainly does to me), but there are two reasons to take it seriously. First, it is how these things will be used practically by the vast majority of people. They are already replacing standard Google search results, and I’ve had many discussions end with people delegating final authority on the truth to an AI. There are…

> people delegating final authority on the truth to an AI I can't avoid looking down on this attitude. But speaks more about people than it speaks about AI: those people want to win an argument, nothing more and nothing else > There are already organizations taking advantage of this, actively producing "AI propaganda", meaning propaganda aimed at the LLMs themselves in order to influence their understanding of what i…

[dead]

Re: Truth is not a direction: a Tarski attack on LLM probes

#86

Post by elendilm »

ziofill wrote: post #74
californical wrote: post #34

Earlier quoted context omitted.

Every breath you take, there is a 99.99% chance that everything is normal, and a 0.01% chance that you breathe mild acid which horribly burns and causes a massive coughing fit. It probably don’t cause long term damage unless that breath happened to be more important than normal, like while driving right as a child runs into the road. Would you act differently knowing that you had one of these occasional acid breaths?…

You are correct, but this is a made up example, not a practical matter. You’re falling into the same trap as the author.

You must have come to rescue him :) from the "incorrect" trap.

Re: Truth is not a direction: a Tarski attack on LLM probes

#87

Post by galaxyLogic »

erehweb wrote: post #20
galaxyLogic wrote: post #11

Earlier quoted context omitted.

Interesting point about "self-referential sentences". I tend to agree. In my view a sentence saying something like "This sentence ..." does not have valid semantic meaning. It says nothing because, what "This" in "This sentence" means is ill-defined. If terms we use are not well-defined, then sentences using such terms can not have meaning. But for the sake of argument let's explore, what could the "this" in (so call…

As the article notes, the sentence "This sentence is written in English" is well understood, and true. "This sentence is written in French" is also well understood and false.

Note, that "This sentence is written in English" is not about the meaning of a sentence but about its encoding. Statements about Truth are about the meaning of something.

What does "This" in the above sentence refer to? It refers to the expression of an idea, how the idea is written, in what language. It does not refer to the idea, or the meaning of the sentence, what we think of as the "sentence" (vs. its encoding). Therefore I would say it is not self-referential sentence, or if it is in a very different way than "This" in "This sentence is false".

Re: Truth is not a direction: a Tarski attack on LLM probes

#88

Post by igleria »

83642736392 wrote: post #85
igleria wrote: post #37

Earlier quoted context omitted.

> people delegating final authority on the truth to an AI I can't avoid looking down on this attitude. But speaks more about people than it speaks about AI: those people want to win an argument, nothing more and nothing else > There are already organizations taking advantage of this, actively producing "AI propaganda", meaning propaganda aimed at the LLMs themselves in order to influence their understanding of what i…

[dead]

If you are asking my opinion, all models are just equally barf inducing to me. It's just that grok also gives me the runs.
Post reply on HN