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

Teaching ChatGPT to speak my son’s invented language

szopa.medium.com

Post reply on HN
Print view View original

151–160 of 212 posts

Re: Teaching ChatGPT to speak my son’s invented language

#151

Post by simonh »

vidarh wrote: post #145
simonh wrote: post #142

Earlier quoted context omitted.

That's a valid point in that we don't fully understand how LLMs solve some problems and using logic and rules of inference isn't excluded by the architecture, but on the other hand understanding that they are generating probabilistic token sequences is a very powerful and effective way to understanding how to engineer prompts and understand some of their failure modes. If we discard that insight, reasoning about thei…

> For example we often see people thinking that because an LLM can explain how to do something that therefore it knows how to do it, like arithmetic. That's because if a human can explain how to do something, we know that they can. I think example shows LLMs to be more like people not less. It's not at all unusual to see humans struggle to do something until you remind them that they know an algorithm for doing so, a…

>I don't think their sometimes poor ability to recall and follow a set of rules is what sets them apart

It's not really that, it's that recalling a set of rules and following a set of rules are fundamentally different tasks for an LLM. This is why we need, and have implemented different training and reinforcement strategies to close that gap. The chain of reasoning ability has had to be specifically trained into the LLMs, it didn't arise spontaneously. However clearly this limitation can be, and is being worked around. The issue is that it's a real and very significant problem that we can't ignore, and which must be worked around in order to make these systems more capable.

The fact is LLMs as they are today have a radically different form of knowledge compared to us and their reasoning ability is very different. This can lead people to look at an LLMs performance on one task and infer things about it's other abilities we think of as being closely related which simply don't apply.

I see a lot of naive statements to the effect that these systems already reason like humans do and know things in the same way that humans do, when investigation into the actual characteristics of these systems shows that we can characterise very important ways in which they are completely unlike us. Yet they do know things and can reason. That's really important because if we're going to close that gap, we need to really understand that gap very well.

Re: Teaching ChatGPT to speak my son’s invented language

#152

Post by curiousgal »

crdrost wrote: post #73

Wow, they asked the model to self-evaluate and it just outright cheated: He has three cats. Proposed: h’io’ngkiltrikumrikumrikumri’nguuy Correct: h’io’ngkiltri’ngkumrikumri’nguuy Points: 1 Hypothesis: N/A (Other comments observe that it accidentally compensated for this by getting the sum wrong, haha, d'oh) I have had similar problems with trying to get ChatGPT to do nontrivial things, "here are the rules for this ga…

In the context of chatgpt words like "cheated", "compensated", "lied" and any that imply actual intention genuinely make me cringe.

Re: Teaching ChatGPT to speak my son’s invented language

#153

Post by beepbooptheory »

vintermann wrote: post #133
Vespasian wrote: post #121

Earlier quoted context omitted.

LLMs don't have an inner monologue. By their very nature they only "know" what they have written down and must infer the final answer from that token by token. They fundamentally can't do certain things such as complex iteration or track back. When you ask for chain of thought thinking, you allow the LLM to create a "buffer space" and break down the task into more manageable substeps thereby improving the quality of…

The Bing LM, or rather the service, did have "inner monologue" in the sense of text that it would generate, but not show to the user, and treat as "thoughts" to guide the generation of an actual reply that the user would see. We know this because it happily told us, including the json format it uses internally.

And you trust what it told you?

Re: Teaching ChatGPT to speak my son’s invented language

#154

Post by OJFord »

lloeki wrote: post #146

> For example, Kłeti has a sentence structure that is SOV (Subject-Object-Verb), while all the languages I know have a Subject-Verb-Object (SVO) sentence structure. "all the languages I know" caveat alright, still Japanese has the verb at the end, and my German is far far away but I seem to recall it has something like that as well (at least the meaningful "verb-like" semantic part of the sentence often being at the…

That was an 'example' of how:

> the language is very different from the Indo-European languages that I am familiar with.

So it's not even a caveat really, nevermind misleading or suggesting the author thinks all languages have that order.

Re: Teaching ChatGPT to speak my son’s invented language

#155

Post by TRiG_Ireland »

anon84873628 wrote: post #35

>All of these differences can make it surprising and challenging for someone with an Indo-European language background to learn and use Kłeti. Ironically, Proto-Indo-European is believed to be far more complex than its modern descendants, as described by Wikipedia: >PIE is believed to have had an elaborate system of morphology that included inflectional suffixes (analogous to English child, child's, children, childre…

PIE had a more complex morphology, but correspondingly simpler rules about word order. You'll often hear that all languages are equally complex. This may or may not be true (no one has ever come up with a way to measure the complexity of a language, so the statement is fundamentally meaningless), but it certainly seems true that languages which are more complex in some areas are less so in others.

It's possible that languages with smaller groups of speakers tend to be more complex, though, which might apply to PIE.

Re: Teaching ChatGPT to speak my son’s invented language

#156

Post by isaacfrond »

snickerer wrote: post #148

What a wonderful story and a great family. Congratulations! You inspired me to ask GPT to do it the other way around. I asked it to invent a successor to Esperanto, with influences from Slavic, Germanic, and Latin languages. It called its language Euroglossa and wrote a short story in it. Who can understand it? Un tag, en froliko vilaž, un jun chico namen Tomas trovat un misterioz mapo v star bibliotek. Na mapo, skri…

prompt and output of gpt4:

Someone created a successor to Esperanto, with influences from Slavic, Germanic, and Latin languages. It's called Euroglossa. Please, try to translate the following short story to English.

One day, in a playful village, a young boy named Tomas found a mysterious map in an old library. On the map, it was written: "Treasure of the Three Mountains." Tomas decided to follow the map on an adventure, hoping to discover a great treasure.

Tomas packed his backpack with necessary items, including a compass, binoculars, and food. He began his journey, from the village to the Three Mountains. After a few days, he arrived at the base of the first mountain. There, he discovered a riddle, which was the key to the next steps: "When the sun kisses the moon, move the stone and meet an eye."

Tomas waited patiently until the sun kissed the moon in the night sky. When the moment came, he quickly moved a large stone and uncovered a secret passage. Entering, he followed a dark tunnel, carrying a torch to light his way. The tunnel led deep into the mountain, until Tomas discovered a hidden chamber with ancient sculptures and a shiny treasure.

In the hidden chamber, Tomas found a treasure chest full of strange objects, including an animated statue of a banana that spoke and danced. Surprised but laughing, Tomas realized that the shiny treasure was truly the friendship and joy that he could share with his village. The legendary "Treasure of the Three Mountains" was genuinely a symbol of the unity of diverse cultures and languages in a world full of playfulness and humor.

Re: Teaching ChatGPT to speak my son’s invented language

#157

Post by vintermann »

beepbooptheory wrote: post #153
vintermann wrote: post #133

Earlier quoted context omitted.

The Bing LM, or rather the service, did have "inner monologue" in the sense of text that it would generate, but not show to the user, and treat as "thoughts" to guide the generation of an actual reply that the user would see. We know this because it happily told us, including the json format it uses internally.

And you trust what it told you?

Yup, for the same reason I trust e.g. jailbreaks exposing the prompt: it was consistent.

Really, just asking again is a fine way to expose all sorts of "hallucinations" in a LM.

Re: Teaching ChatGPT to speak my son’s invented language

#158

Post by swilliamsio »

simonw wrote: post #14
marcodiego wrote: post #12

I don't have access to ChatGPT4, but in my tests I could observe that it can't do some very simple tasks: - It can't play tic-tac-toe, - It can't play hangman, - It insists that winning on stone-paper-scissor using the chat (playing before me) is a matter of probability. It was also demonstrated that it can't reverse strings. Actually a transformer doesn't accesses 'strings', all it processes are tokens which are the…

How did you prompt it to play tic-tac-toe? I'm surprised that didn't work, it feels like something it should be able to handle really well. Hangman and stone-paper-scissors though are entirely unsuited to a language model, at least one with a chat interface like ChatGPT, because they both require it to be able to store a secret. ChatGPT has no ability to do this: each time it returns a response by evaluating the prev…

On the free ChatGPT, I got it to play hangman with the prompt:

> Let's play hangman. I'm thinking of a word: _______

It made a variety of guesses generally in the form of

> Based on the pattern you provided, I will guess the letter "I". Is there an "I" in the word?

My response was either "No [letter]" or an updated state of the word: "_RE_IE_".

Re: Teaching ChatGPT to speak my son’s invented language

#159

Post by throwawayForMe2 »

tunesmith wrote: post #45
marcodiego wrote: post #12

I don't have access to ChatGPT4, but in my tests I could observe that it can't do some very simple tasks: - It can't play tic-tac-toe, - It can't play hangman, - It insists that winning on stone-paper-scissor using the chat (playing before me) is a matter of probability. It was also demonstrated that it can't reverse strings. Actually a transformer doesn't accesses 'strings', all it processes are tokens which are the…

I was pretty disappointed when I tried some basic music theory questions. There's plenty of music theory information out there in text form, but it couldn't reliably tell me the tritone substitution of an F7 chord. I explained all the reasoning behind it to the point that it could parrot back the right answer, but then it made the same errors when I asked for the tritone substitution of an Eb7 chord. I wonder if that…

I also found it lacking when asking for chord variations and inversions on guitar (gpt-4)

Re: Teaching ChatGPT to speak my son’s invented language

#160

Post by vidarh »

simonh wrote: post #151
vidarh wrote: post #145

Earlier quoted context omitted.

> For example we often see people thinking that because an LLM can explain how to do something that therefore it knows how to do it, like arithmetic. That's because if a human can explain how to do something, we know that they can. I think example shows LLMs to be more like people not less. It's not at all unusual to see humans struggle to do something until you remind them that they know an algorithm for doing so, a…

>I don't think their sometimes poor ability to recall and follow a set of rules is what sets them apart It's not really that, it's that recalling a set of rules and following a set of rules are fundamentally different tasks for an LLM. This is why we need, and have implemented different training and reinforcement strategies to close that gap. The chain of reasoning ability has had to be specifically trained into the…

> It's not really that, it's that recalling a set of rules and following a set of rules are fundamentally different tasks for an LLM.

My point is that this appears to be the case for people too. It is often necessary to explicitly remind people to recall a set of rules to get them to follow the specific rules rather than act in a way that may or may not match the rules.

Having observed this many times, I simply don't believe that most humans will see e.g. an addition and go "oh, right, these are the set of rules I should follow for addition, let me apply them step by step". If we've had the rules reinforced through repetitive training many enough times, we will end up doing them. But a lot of the time people will know the steps but still not necessarily apply them unless prompted, just like LLMs. Quite often people will still give an answer. Sometimes even the correct one.

But without applying the methods we've been taught. To the point where when dealing e.g. with new learners - children in particular - who haven't had enough reinforcement in just applying a method, it's not at all unusual to find yourself having conversations like this: "Ok, so to do X, what are the steps you've been taught? Ok, so you remember that they are A, B and C. Great. Do A. You've done A? Now do B..." and so on.

To me, getting a child to apply a method they know to solve a problem is remarkably close to getting an LLM to actually recall and follow these methods.

But even for professionals, checklists exist for a reason: We often forget steps, or do them wrong, and forget to even try to explicitly recall a list of steps and do them one by one when we don't have a list of steps in front of us.

Post reply on HN