site banner
naraburns  nihil supernum  2yr ago (text post)   35342 thread views

Culture War Roundup for the week of March 13, 2023

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

15
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Using the model here using it's GRE scores.

GPT-3.5 Verbal IQ = 118

GPT-3.5 Quantitative IQ = 101

GPT-4 Verbal IQ = 144

GPT-4 Quantitative IQ = 130

It's inconsistent though. It completely bombed the AMC 10/12

I'd say the AMC is harder than the GRE where it counts (fluid g).

Of course the AMC is harder, but this did about as poor as random guessing on the AMC 10. That's much worse than "quantitative IQ 130" level.

Less study material in its dataset I should think -- you would do well on the GRE too if you memorized every study guide on the internet.

Then that suggests that it's GRE score is not an effective measure of its "IQ"

I have a vested interest in the GRE being an effective measure because it would make me rather high IQ lol.

So as a defense of the GRE I will let it be known that ETS (GRE test makers) hires a suspiciously large number of psychometric PhDs, the last time I checked more than half the job openings were for psychometricians, and they know exactly what they are doing (making a socially acceptable IQ test). If its bad at that, it's probably not for a lack of trying.

And that an "IQ test" a language model has an advantage on, probably because of plenty of training data doesn't imply humans are prone to that failure (success) mode of the test. Not for a difference in kind but magnitude, no human reads literal billions of tokens.

I don't know what the implications/inferences are. It's certainly interesting that a LLM can do non first order quantitative reasoning questions at all to me. Suggests to me that is there an overlap of language and quantitative reasoning in whatsoever space GPT is pulling its inferences from, might even be universal.

Sure -- same goes for the tests that it did score well on though.

Ed: sorry misread -- yes, that's exactly what I'm suggesting. All of the tests they are giving it are testing its ability to memorize study material from its training set -- which may be useful for some things, but in no sense is "intelligence", particularly not "intelligence" as in "Artificial General Intelligence"

Ideally one could test this by writing a test of pretty low difficulty level (say tenth grade non-advanced math) but with questions framed in a completely different way from the AP/SAT type stuff in the dataset. Then compare results with an actual tenth grader.

There's an updated version of that here on which the score is even higher. Although there is probably plenty of online info on standardized tests and whatnot in its training set, if it's from scrubbing the internet, so I doubt you can infer a ton about how "smart" it is in general from these. The revised conversion yields Verbal IQ=146, Quant IQ=135.5, overall IQ=145.4.

Although there is probably plenty of online info on standardized tests and whatnot in its training set, if it's from scrubbing the internet, so I doubt you can infer a ton about how "smart" it is in general from these.

This has been an annoying aspect of LLM AI hype. There are plenty of indicators of something going on but many of the test results are not of that set. If you train them on the question sets and answer keys for repeatably mechanically gradable exams like the SAT, GRE or bar exams then it should be expected that they will perform well on them.

What would be really nice is if whenever the AI produced content it had to also tell you the minimal edit distance between that content and some content (or a direct combination of contents) in its training set. That way you could have a good measure of how much original content it was actually producing vs. how much it was just paraphrasing its training set. Or at least it would be useful to have extensive data on the average edit distance between a response and some item in the corpus.

Your proposed method doesn't work - even if you just turned a query into an embedding, picked the closest text in its dataset, and then ran that text through google translate and back a few times to obfuscate word choice, order, and other things like that, it'd change enough that literal edit distance would still be very high.

An analogy to image models, here's a claimed example of of taking inspiration from a particular photo in a training set. It's really not that close.

Not really, no. Edit distance is relative to which operations count as primitive "edits" and the "cost" of each use of that operation. There are specific forms of edit distance of which what you are saying is true, but you could also have an edit distance where "run it through Google Translate" is a primitive edit operation. Obviously, you would have to pick the operations to fit the specific model, e.g. what external resources it has access to.

Okay, I assumed you meant character-level edit distance, because that's what the article you linked was exclusively about. But without that, 'edit distance' isn't really a useful term, as we don't know what a 'primitive operation' is in the context of a LLM because we do not know that much if anything about what they actually do internally.

Based on what I've seen, 'most similar example in training set' doesn't capture the extent to which LLMs memorize things. Even if they are memorizing a lot, it's memorization in a very complicated way - otherwise the 'write a story about X in the style of Y', where X and Y hadn't ever been done before, just wouldn't work.

how is this memorizing? or this? Like, there's some extent to which it's memorizing things more than humans do, certainly. But it's a very vague sense, and positing a metric like edit distance that fully captures that sense just restates the problem, because we don't know what that is

  1. That isn’t true, the formal definition doesn’t restrict what operations there are (if you disagree then you should quote it and tell me where), it just requires that they be operations on strings. Whatever else using Google Translate is, it’s an operation on strings.

  2. I’m not suggesting that we base primitive operations on what the LLM does internally. Your initial example wasn’t about what the LLM does internally, it was about “what if the LLM ran things through Google Translate a few times to trick you.” I’m saying you could supplement a more basic measure with additional operations to capture when the model has access to external programs like Google Translate.

  3. I never said that the AI was just “memorizing” things, that’s an obvious strawman. All I said was that edit distance would help you tell how much the AI was paraphrasing from its training set, which doesn’t presuppose that it’s not paraphrasing very little. You seem to have misinterpreted me as saying that an edit distance would show the AI was paraphrasing a lot, when all I said was an edit distance would be a helpful metric of how much it was. I didn’t make any specific commitments about how much that would be across a wide range of cases. Nor did I suggest that an edit distance would fully capture things, just that it would give a good sense of how close things were.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats