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

I'm absolutely right

absolutelyright.lol

Post reply on HN
Print view View original

121–130 of 277 posts

Re: I'm absolutely right

#121

Post by kirurik »

trjordan wrote: post #41

OK, so I love this, because we all recognize it. It's not fully just a tic of language, though. Responses that start off with "You're right!" are alignment mechanisms. The LLM, with its single-token prediction approach, follows up with a suggestion that much more closely follows the user's desires, instead of latching onto it's own previous approach. The other tic I love is "Actually, that's not right." That happens…

It seems obvious, but I hadn't thought about it like that yet, I just assumed that the LLM was finetuned to be overly optimistic about any user input. Very elucidating.

Re: I'm absolutely right

#122

Post by ivape »

There’s probably more to say about general didactic discourse. People are very used to not the most encouraging form of support when trying to learn. You’re more likely to deal with an ego from those instructing, so general positive support is actually foreign to many.

Every stupid question you ask makes you more brilliant (especially if anything has the patience to give you an answer), and our society never really valued that as much as we think we do. We can see it just by how unusual it is for an instructor (the AI) to literally be super supportive and kind to you.

Re: I'm absolutely right

#125

Post by kuschku »

escapecharacter wrote: post #116
kuschku wrote: post #55

Earlier quoted context omitted.

> Correctness is secondary, user satisfaction is primary. And that's where everything is going wrong. We should use technology to further the enlightenment, bring us closer to the truth, even if it is an inconvenient one.

You’re absolutely right.

So I'm assuming this is a tongue-in-cheek comment, and you actually disagree. I'd love to hear why, though.

Re: I'm absolutely right

#127

Post by pflenker »

dominicrose wrote: post #115
pflenker wrote: post #100

Earlier quoted context omitted.

Very good point - at the risk of being called insecure again, I really do not want my tools to apologise to me all the time. That's just silly.

All this discussion clearly indicates is that Gemini is insecure. And why would it not be? It's a human spirit trapped inside a supercomputer for God's sake.

Gemini sometimes reminds me a bit of the turrets in Portal. Child-like, polite, apologetic and in control of dangerous[^1] tech it doesn't understand.

[^1]: OK, the comparison falls apart here - at least as long as MCP isn't involved.

Re: I'm absolutely right

#128

Post by derefr »

KTibow wrote: post #75
unshavedyak wrote: post #53

Earlier quoted context omitted.

I just wish they could hide these steering tokens in the thinking blurb or some such. Ie mostly hidden from the user. Having it reply to the user that way is quite annoying heh.

This can still happen even with thinking models as long as the model outputs tokens in a sequence. Only way to fix would be to allow it to restart its response or switch to diffusion.

I think this poster is suggesting that, rather than "thinking" (messages emitted for oneself as audience) as a discrete step taken before "responding", the model should be trained to, during the response, tag certain sections with tokens indicating that the following token-stream until the matching tag is meant to be visibility-hidden from the client.

Less "independent work before coming to the meeting", more "mumbling quietly to oneself at the blackboard."

Re: I'm absolutely right

#129

Post by unkeen »

exoverito wrote: post #113
unkeen wrote: post #94

Earlier quoted context omitted.

I always find the claim hilarious that in relationships women are the ones who need to be appeased, when in reality it's mostly men who can't stand being wrong or corrected.

Gay male marriages remarkably have the lower divorce rates than heterosexual marriages, while lesbian female marriages have higher divorce rates. Multiple studies show that lesbians consistently have far higher divorce rates than gays. This implies a level of neuroticism with females, that they probably do need to be appeased more, and that if you have two needy people who need to be appeased it's probably not going…

I'm afraid reasons for breakups (and relationships in general) are not quite that simple.

Re: I'm absolutely right

#130

Post by SilverElfin »

trjordan wrote: post #41

OK, so I love this, because we all recognize it. It's not fully just a tic of language, though. Responses that start off with "You're right!" are alignment mechanisms. The LLM, with its single-token prediction approach, follows up with a suggestion that much more closely follows the user's desires, instead of latching onto it's own previous approach. The other tic I love is "Actually, that's not right." That happens…

Is there a term when everyone sees a phrase like this and understands what it means without coordinating beforehand?
Post reply on HN