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

AI hallucinations: Why LLMs make things up (and how to fix it)

kapa.ai

Post reply on HN
Print view View original

231–240 of 257 posts

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#231

Post by mrguyorama »

GuB-42 wrote: post #157
lolinder wrote: post #8

> While the hallucination problem in LLMs is inevitable [0], they can be significantly reduced... Every article on hallucinations needs to start with this fact until we've hammered that into every "AI Engineer"'s head. Hallucinations are not a bug—they're not a different mode of operation, they're not a logic error. They're not even really a distinct kind of output. What they are is a value judgement we assign to the…

Of course they are a bug. Just that hallucination emerge from the normal function of a LLM doesn't make it "not a bug". No programmer in their right mind will call the lack of bound checking resulting in garbled output "not a bug", even though it is a totally normal thing to do from the point of view of a CPU. It is a bug and you need additional code to fix it, for example by checking for out-of-bounds condition and…

>Of course they are a bug

No.

When you build a bloom filter and it says "X is in the set" and X is NOT in the set, that's not a bug, that's an inherent behavior of the very theory of a probabilistic data structure. It is something that WILL happen, that you MUST expect to happen, and you MUST build around.

>And to fix it, we need to engineer solutions that prevent the hallucinations from happening

The whole point is that this is fundamentally impossible.

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#232

Post by mdp2021 »

PittleyDunkin wrote: post #204
tsujamin wrote: post #182

Earlier quoted context omitted.

I’d argue hallucinations are unexpected in LLMs by the large (non technical) number of users who use them directly, or indirectly though other services. It all depends on whose specification you’re assessing the “bugginess” against, the inference code as written, the research paper, colloquial understanding in technical circles, or how the product is pitched and presents to users.

> I’d argue hallucinations are unexpected in LLMs by the large (non technical) number of users who use them directly, or indirectly though other services. People also blithely trust other humans even against all evidence that they're trustworthy. Some things just aren't fixable.

The median individual is _not_ a model, and cannot represent the whole of the set. If the median is incompetent, the competent remain competent.

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#233

Post by drewbeck »

SirMaster wrote: post #223
stonemetal12 wrote: post #207

Earlier quoted context omitted.

> LLMs naturally hallucinate, but it is not what we want, so it is a bug. I rolled a one in D&D, it is not what I wanted, so it is a bug. Remove it from all my dice.

What? You are telling me that when you roll a 6 sided dice you are not expecting any of the 1-6 as a result? If a 6-sided dice produced a 7 that would be a bug. When you rolled a dice, I would argue that you knew you wanted a random number from 1-6, not that you wanted a specific number or not a specific number. If you wanted that you wouldn't have used a dice. When I ask an LLM to write code for me and it references…

IMO it is because you just asked a bunch of dice to write code for you.

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#234

Post by mdp2021 »

butlike wrote: post #188
ToucanLoucan wrote: post #181

Earlier quoted context omitted.

Except we put up with network lag because it's an understandable, if undesirable, caveat to an otherwise useful technology. No one would ever say that because a network is sometimes slow, that it is then preferable to not have computers networked. The benefits clearly outweigh the drawbacks. This is not true for many applications of LLM. Generating legal documents, for example: it is not acceptable that it hallucinat…

If it's not acceptable to hallucinate laws for writing legal documents, then writing legal documents is probably an unacceptable use case. Also, how do you mitigate a lawyer writing whatever they want (aka: hallucinating) when writing legal documents? Double-checking??

> Also, how do you mitigate a lawyer writing whatever they want (aka: hallucinating) when writing legal documents? Double-checking??

Of course they are supposed to double and triple and multiple check as they think and write, documentation and references at hand, _exactly_ how you are supposed to do from trivial informal context on towards critical ones - exactly the same, you check the detail and the whole, multiple times.

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#235

Post by mdp2021 »

sdesol wrote: post #230
mdaniel wrote: post #213

Earlier quoted context omitted.

Wow, that sounds great: just have every customer who interacts with your LLM come back to the site in 2 days to get the real answer to their question. How can I invest?

I've said before, but I'm not convinced LLM should be public facing. I know some companies have been burned by them and in my opinion, LLM should be about helping customer support people find answers faster.

> LLM should be about helping customer support people find answers faster

That would be as dangerous as any other function: you still need personnel verified as trustworthy in processing unreliable input.

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#236

Post by mdp2021 »

zmgsabst wrote: post #193
mort96 wrote: post #126

Earlier quoted context omitted.

Healthy humans generally have some internal model of the world against which they can judge what they're about to say. They can introspect and determine whether what they say is a guess or a statement of fact. LLMs can't.

Humans routinely misremember facts but are relatively certain those remembrances are correct. That’s a form of minor, everyday hallucination. If you engage in such thorough criticism and checking of every recalled fact as to eliminate that, you’ll crush your ability to synthesize or compose new content.

> If you engage in such thorough criticism and checking of every recalled fact as to eliminate that, you’ll

Experience tells us differently: creativity is not impacted. In fact, it will probably return better solutions (as opposed to delirious).

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#237

Post by mdp2021 »

Terr_ wrote: post #3

When people talk about stopping an LLM from "seeing hallucinations instead of the truth", that's like stopping an Ouija-board from "channeling the wrong spirits instead of the right spirits." It suggests a qualitative difference between desirable and undesirable operation that isn't really there. They're all hallucinations, we just happen to like some of them more than others.

> It suggests a qualitative difference

And what is sought is, in a way, a jump to that qualitative difference. (And surely there are «desirable and undesirable operation[s]».)

"Add something to the dices so that they can be well predictive".

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#238

Post by mdp2021 »

skydhash wrote: post #150
IWeldMelons wrote: post #147

Earlier quoted context omitted.

Not as interactive, not gamified.

School, then. Which is so gamified that it has real stakes. And so much interactions.

With an interactive LLM you can take a manual and start asking questions (about what you read), also recursively.

It is a very efficient way of studying. No, doing it with a professor is not the same - unless you can afford an always available tutor of unthinkable erudition.

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#239

Post by drcwpl »

lolinder wrote: post #8

> While the hallucination problem in LLMs is inevitable [0], they can be significantly reduced... Every article on hallucinations needs to start with this fact until we've hammered that into every "AI Engineer"'s head. Hallucinations are not a bug—they're not a different mode of operation, they're not a logic error. They're not even really a distinct kind of output. What they are is a value judgement we assign to the…

You are right - "Hallucinations are not a bug"

Re: AI hallucinations: Why LLMs make things up (and how to fix it)

#240

Post by mdp2021 »

imchillyb wrote: post #161

Toddlers don't understand truth either, until it's taught. This crayon is red. This crayon is blue. The adult asks: "is this crayon red?" The child responds: "no that crayon is blue." The adult then affirms or corrects the response. This occurs over and over and over until that child understands the difference between red and blue, orange and green, yellow and black etcetera. We then move on to more complex items and…

You probably need to be more clear: the LLM is trained with large amounts of data making statements about facts. It is told repeatedly, "according to this source that crayon is blue".
Post reply on HN