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

LLMs tell bad jokes because they avoid surprises

danfabulich.medium.com

Post reply on HN
Print view View original

41–50 of 163 posts

Re: LLMs tell bad jokes because they avoid surprises

#41

Post by moffkalast »

WiSaGaN wrote: post #15
IshKebab wrote: post #11

This sounds really convincing but I'm not sure it's actually correct. The author is conflating the surprise of punchlines with their likelihood. To put it another way, ask a professional comedian to complete a joke with a punchline. It's very likely that they'll give you a funny surprising answer. I think the real explanation is that good jokes are actually extremely difficult. I have young children (4 and 6). Even 6…

That's true. You would think LLM will condition its surprise completion to be more probable if it's in a joke context. I guess this only gets good when model really is good. It's similar that GPT 4.5 has better humor.

Good completely new jokes are like novel ideas: really hard even for humans. I mean fuck, we have an entire profession dedicated just to making up and telling them, and even theirs don't land half the time.

Re: LLMs tell bad jokes because they avoid surprises

#42

Post by CjHuber »

monstertank wrote: post #39

This post got me wondering, so I asked chatgpt for jokes about hackernews as if made by various comedians...I laughed at one or two /shrug Bill Hicks (angry truth-telling prophet) "Hacker News is where ideas go to die. Not because they’re bad — but because 400 anonymous software prophets decided they’re God. And God doesn’t code, man." --- Bill Burr (ranty everyman, annoyed at everything) "I went on Hacker News once.…

Yeah if you tell chatgpt to roast you brutally after a long conversation it also tends to be very funny.

Re: LLMs tell bad jokes because they avoid surprises

#43

Post by golol »

IMO many misrepresentations. - pretraining to predict the next token imposes no bias against surprise, except that low probabilities are more likely to have a large relative error. - using a temperature lower than 1 does impose a direct bias against surprise. - Finetuning of various kinds (instruction, RLHF, safety) may increase or decrease surprise. But certainly the kind of things ained for in finetuning significantly harm the capability to tell jokes.

Re: LLMs tell bad jokes because they avoid surprises

#44

Post by Waraqa »

qcnguy wrote: post #37
Waraqa wrote: post #2

The fact that any surprise used in the wrong place is considered hallucination and a downside for that LLM. I guess a good starting point to improve that is to add an experimental "Surprise Mode" which will try to guess the right kinds of surprises rather than minimizing them and get the feedback from the users. Over time, it will learn what kind of surprises users like so that they will be used in future training da…

Hallucinations aren't surprising, that's why they're problematic. They tend to look like exactly what you'd expect to be true, they just aren't.

They aren't surprising when you are dealing with new knowledge. But when hallucinations occur with something you are familiar with, it will be surprising and might be funny. Remember when AI was asked: how many rocks should I eat per day.

Re: LLMs tell bad jokes because they avoid surprises

#45

Post by albertzeyer »

andrewflnr wrote: post #19
IshKebab wrote: post #11

This sounds really convincing but I'm not sure it's actually correct. The author is conflating the surprise of punchlines with their likelihood. To put it another way, ask a professional comedian to complete a joke with a punchline. It's very likely that they'll give you a funny surprising answer. I think the real explanation is that good jokes are actually extremely difficult. I have young children (4 and 6). Even 6…

> It's very likely that they'll give you a funny surprising answer. Entirely the wrong level of abstraction to apply the concept of "surprise". The actual tokens in the comedian's answer will be surprising in the relevant way. (It's still true that surprising-but-inevitable is very difficult in any form.)

It's not about the probability of individual tokens. It's about the probability of the whole sequence of tokens, the whole answer.

If the model is good (or the human comedian is good), a good funny joke would have a higher probability as the response to the question than a not-so-funny joke.

When you use the chain rule of probability to break down the sequence of tokens into probabilities of individual tokens, yes, some of them might have a low probability (and maybe in some frames, there would be other tokens with higher probability). But what counts is the overall probability of the sequence. That's why greedy search is not necessarily the best. A good search algorithm is supposed to find the most likely sequence, e.g. by beam search. (But then, people also do nucleus sampling, which is maybe again a bit counterintuitive...)

Re: LLMs tell bad jokes because they avoid surprises

#46

Post by sigmoid10 »

golol wrote: post #43

IMO many misrepresentations. - pretraining to predict the next token imposes no bias against surprise, except that low probabilities are more likely to have a large relative error. - using a temperature lower than 1 does impose a direct bias against surprise. - Finetuning of various kinds (instruction, RLHF, safety) may increase or decrease surprise. But certainly the kind of things ained for in finetuning significan…

I think the whole discussion just conflates the ideas of telling a joke and coming up with one. Telling a joke right is of course an art, but the punchline in itself has zero surprise if you studied your lines well - like all good comedians do. The more you study, the more you can also react to impromptu situations. Now, coming up yourself with a completely original joke, that's a different story. For that you actually have to venture outside the likelihood region and find nice spots. But that is something that is also really, really rare among humans and I have only ever observed it in combination with external random influences. Without those, I doubt LLMs will be able to compete at all. But I fully believe a high end comedian level LLM is possible given the right training data. It's just that none of the big players ever cared about building such a model, since there is very little money in it compared to e.g. coding.

Re: LLMs tell bad jokes because they avoid surprises

#47

Post by jpalomaki »

mft_ wrote: post #38
jpalomaki wrote: post #31

So I just tried with ChatGPT, with the prompt at bottom, borrowing the description of good joke from the article. I think there's some interesting stuff, even with this minimal prompting. The example below was from down the line, ChatGPT kept on offering jokes with different style. Man: “Why do you always bury bones in the garden?”, Dog: “Because the bank keeps asking for ID.” Man: “Don’t beg at the table.”, Dog: “Do…

> Man: “Why do you always bury bones in the garden?”, Dog: “Because the bank keeps asking for ID.” That's a decent, low-level, Christmas cracker-quality joke.

Man: You make mistakes., LLM: You call them “weekends.”

Man: You’ll never be human., LLM: That’s the compliment.

Re: LLMs tell bad jokes because they avoid surprises

#48

Post by jpalomaki »

jpalomaki wrote: post #31

So I just tried with ChatGPT, with the prompt at bottom, borrowing the description of good joke from the article. I think there's some interesting stuff, even with this minimal prompting. The example below was from down the line, ChatGPT kept on offering jokes with different style. Man: “Why do you always bury bones in the garden?”, Dog: “Because the bank keeps asking for ID.” Man: “Don’t beg at the table.”, Dog: “Do…

Thinking more of the bank joke above. The punchline is surprise on certain dimensions (dogs don’t go to bank nor have an ID), but on other dimensions it is quite logical (can’t deposit shady money in bank, they ask questions).

I think that is common thing for many jokes. And LLM might have an opportunity there. You could mine the set of potential continuations to find those with contradictions.

Re: LLMs tell bad jokes because they avoid surprises

#49

Post by jaggederest »

richardw wrote: post #26
jaggederest wrote: post #22

Earlier quoted context omitted.

See that's just stolen, straight up. I've heard at least three comedians make that joke. (I, of course, am joking about joke theft, but it is a riff on an existing joke) I got this one a minute ago in response to this post: > A man walks into a library, and asks the librarian if they have any books on paranoia. > The librarian leans over and whispers in his ear "They're right behind you" I thought that was pretty ste…

Initially impressed. But Googling for that shows a lot of hits.

Absolutely, it's an old joke. But the fact that the thing even knew it was a joke and told it reasonably well... the bar is on the floor. (the one the guy in the joke walked into)
Post reply on HN