Claude is good at assembling blocks, but still falls apart at creating them
approachwithalacrity.com261 points by bblcla 2 days ago
261 points by bblcla 2 days ago
I've yet to be convinced by any article, including this one, that attempts to draw boxes around what coding agents are and aren't good at in a way that is robust on a 6 to 12 month horizon.
I agree that the examples listed here are relatable, and I've seen similar in my uses of various coding harnesses, including, to some degree, ones driven by opus 4.5. But my general experience with using LLMs for development over the last few years has been that:
1. Initially models could at best assemble a simple procedural or compositional sequences of commands or functions to accomplish a basic goal, perhaps meeting tests or type checking, but with no overall coherence,
2. To being able to structure small functions reasonably,
3. To being able to structure large functions reasonably,
4. To being able to structure medium-sized files reasonably,
5. To being able to structure large files, and small multi-file subsystems, somewhat reasonably.
So the idea that they are now falling down on the multi-module or multi-file or multi-microservice level is both not particularly surprising to me and also both not particularly indicative of future performance. There is a hierarchy of scales at which abstraction can be applied, and it seems plausible to me that the march of capability improvement is a continuous push upwards in the scale at which agents can reasonably abstract code.
Alternatively, there could be that there is a legitimate discontinuity here, at which anything resembling current approaches will max out, but I don't see strong evidence for it here.
It feels like a lot of people keep falling into the trap of thinking we’ve hit a plateau, and that they can shift from “aggressively explore and learn the thing” mode to “teach people solid facts” mode.
A week ago Scott Hanselman went on the Stack Overflow podcast to talk about AI-assisted coding. I generally respect that guy a lot, so I tuned in and… well it was kind of jarring. The dude kept saying things in this really confident and didactic (teacherly) tone that were months out of date.
In particular I recall him making the “You’re absolutely right!” joke and asserting that LLMs are generally very sycophantic, and I was like “Ah, I guess he’s still on Claude Code and hasn’t tried Codex with GPT 5”. I haven’t heard an LLM say anything like that since October, and in general I find GPT 5.x to actually be a huge breakthrough in terms of asserting itself when I’m wrong and not flattering my every decision. But that news (which would probably be really valuable to many people listening) wasn’t mentioned on the podcast I guess because neither of the guys had tried Codex recently.
And I can’t say I blame them: It’s really tough to keep up with all the changes but also spend enough time in one place to learn anything deeply. But I think a lot of people who are used to “playing the teacher role” may need to eat a slice of humble pie and get used to speaking in uncertain terms until such a time as this all starts to slow down.
> in general I find GPT 5.x to actually be a huge breakthrough in terms of asserting itself when I’m wrong
That's just a different bias purposefully baked into GPT-5's engineered personality on post-training. It always tries to contradict the user, including the cases where it's confidently wrong, and keeps justifying the wrong result in a funny manner if pressed or argued with (as in, it would have never made that obvious mistake if it wasn't bickering with the user). GPT-5.0 in particular was extremely strongly finetuned to do this. And in longer replies or multiturn convos, it falls into a loop on contradictory behavior far too easily. This is no better than sycophancy. LLMs need an order of magnitude better nuance/calibration/training, this requires human involvement and scales poorly.
Fundamental LLM phenomena (ICL, repetition, serial position biases, consequences of RL-based reasoning etc) haven't really changed, and they're worth studying for a layman to get some intuition. However, they vary a lot model to model due to subtle architectural and training differences, and impossible to keep up because there are so many models and so few benchmarks that measure these phenomena.
By the time I switched to GPT 5 we were already on 5.1, so I can't speak to 5.0. All I can say is that if the answer came down to something like "push the bias in the other direction and hope we land in the right spot"... well, I think they landed somewhere pretty good.
Don't get me wrong, I get a little tired of it ending turns with "if you want me to do X, say the word." But usually X is actually a good or at least reasonable suggestion, so I generally forgive it for that.
To your larger point: I get that a lot of this comes down to choices made about fine tuning and can be easily manipulated. But to me that's fine. I care more about if the resulting model is useful to me than I do about how they got there.
By the time GPT 5.5 landed we were already on 5.1, honestly they seem to converge on similar limitations around compositional reasoning.
> That's just a different bias purposefully baked into GPT-5's engineered personality on post-training.
I want to highlight this realization! Just because a model says something cool, it doesn't mean it's an emergent behavior/realization, but more likely post-training.
My recent experience with claude code cli was exactly this.
It was so hyped here and elsewhere I gave it a try and I'd say it's almost arrogant/petulant.
When I pointed out bugs in long sessions it tried to gaslight me that everything was alright, faked tests to prove his point.
"Still on Claude Code" is a funny statement, given that the industry is agreeing that Anthropic has the lead in software generation while others (OpenAI) are lagging behind or have significant quality issues (Google) in their tooling (not the models). And Anthropic frontier models are generally "You're absolutely right - I apologize. I need to ..." everytime they fuck something up.
Why is it every time anyone has a critique someone has to say “oh but you aren’t using model X, which clearly never has this problem and is far better”?
Yet the data doesn’t show all that much difference between SOTA models. So I have a hard time believing it.
GP here: My problem with a lot of studies and data is that they seem to measure how good LLMs are at a particular task, but often don't account for "how good the LLM is to work with". The latter feels extremely difficult to quantify, but matters a lot when you're having a couple dozen turns of conversation with an LLM over the course of a project.
Like, I think there's definitely value in prompting a dozen LLMs with a detailed description of a CMS you want built with 12 specific features, a unit testing suite and mobile support, and then timing them to see how long they take and grading their results. But that's not how most developers use an LLM in practice.
Until LLMs become reliable one-shot machines, the thing I care most about is how well they augment my problem solving process as I work through a problem with them. I have no earthly idea of how to measure that, and I'm highly skeptical of anyone who claims they do. In the absence of empirical evidence we have to fall back on intuition.
A friend recommended to me having a D&D style roleplay with some different engines, to see which you vibe with. I thought this sounded crazy but I took their advice.
I found this worked suprisingly well, I was certain 'claude' was best, while they like grok and someone else liked ChatGPT. Some AIs just end up fitting best with how you like to chat I think. I do definately also find claude best for coding with as well.
Because the answer to the question, “Does this model work for my use case?” is subjective.
Because they are getting better. They're still far from perfect/AGI/ASI, but when was the last time you saw the word "delve"? So the models are clearly changing, the question is why doesn't the data show That they're actually better?
Thing is, everyone knows the benchmarks are being gamed. Exactly how is besides the point. In practice, anecdotally, Opus 4.5 is noticably better than 4, and GPT 5.2 has also noticably improved. So maybe the real question is why do you believe this data when it seems at odds with observations by humans in the field?
> Jeff Bezos: When the data and the anecdotes disagree, the anecdotes are usually right.
https://articles.data.blog/2024/03/30/jeff-bezos-when-the-da...
The type of person who outsources their thinking to their social media feed news stories and isn't intellectually curious enough to deeply explore the models themselves in order for the models to display their increase in strength, isn't going to be able to tell this themselves.
I would think this also correlates with the type of person who hasn't done enough data analysis themselves to understand all the lies and misleading half-truths "data" often tells. In the reverse also, that experience with data inoculates one to some degree against the bullshitting LLM so it is probably easier to get value from the model.
I would imagine there are all kinds of factors like this that multiple so some people are really having vastly different experiences with the models than others.
"They dont say X as often anymore" is just a distraction, it has nothing to do with actual capability of the model.
Unfortunately, I think that the overlap between actual model improvements and what people perceive as "better" is quite small. Combine this with the fact that most people desperately want to have a strong opinion on stuff even though the factual basis is very weak.. "But I can SEE it is X now".
"They don't use delve anymore" is not really a testament that they became better.
Most of what I can do now with them I could do half a year to a year ago. And all the mistakes and fail loops are still there, across all models.
What changed is the number of magical incantations we throw at these models in the form of "skills" and "plugins" and "tools" hoping that this will solve the issue at hand before the context window overflows.
> I haven’t heard an LLM say anything like that since October, and in general I find GPT 5.x
It said precisely that to me 3 or 4 days ago when I questioned its labelling of algebraic terms (even though it was actually correct).
Claude is still just like that once you’re deep enough in the valley of the conversation. not exactly that phrase but things like that’s the smoking gun or so. nothing has changed.
> Claude is still just like that once you’re deep enough in the valley of the conversation
My experience is claude (but probably other models as well) indeed resort to all sorts of hacks once the conversation has gone for too long.
Not sure if it's an emergent behavior or something done in later stages of training to prevent it from wasting too many tokens when things are clearly not going well.
People desperately want 'the plateau' to be true because it means our jobs would be safe and we could call ourselves experts again. If the ground is keep moving then no one is truly an expert. There is just no enough time to achieve expertise when the paradigm shifts every six months.
I don't see a reason to think we're not going to hit a plateua sooner or later (and probably sooner). You can't scale your way out of hallucinations, and you can't keep raising tens of billions to train these things without investors wanting a return. Once you use up the entire internets worth of stack overflow responses and public github repositories you run into the fact that these things aren't good at doing things outside their training dataset.
Long story short, predicting perpetual growth is also a trap.
> Once you use up the entire internets worth of stack overflow responses and public github repositories you run into the fact that these things aren't good at doing things outside their training dataset.
I think the models have reached that human training data limitation a few generations ago, yet they stil clearly improve by various other techniques.
I agree with a lot of what you've said, but I completely disagree that LLM's are no longer sycophantic. GPT-5 is definitely still very sycophantic, 'You're absolutely right!' still happens, etc. It's true it happens far less in a pure coding context (Claude Code / Codex) but I suspect only because of the system prompts, and those tools are by far in the minority of LLM usage.
I think it's enlightening to open up ChatGPT on the web with no custom instructions and just send a regular request and see the way it responds.
Opus 4.5 seems to be better than GPT 5.2 or 5.2 Codex at using tools and working for long stretches on complex tasks.
I used to get made up APIs in functions, now I get them in modules. I used to get confidently incorrect assertions in files now I get them across codebases.
Hell, I get poorly defined APIs across files and still get them between functions. LLMs aren't good at writing well defined APIs at any level of the stack. They can attempt it at levels of the stack they couldn't a year ago, but they're still terrible at it unless the problem is so well known enough that they can regurgitate well reviewed code.
I still get made-up Python types all the time with Gemini. Really quite distracting when your codebase is massive and triggers a type error, and Gemini says
"To solve it you just need to use WrongType[ThisCannotBeUsedHere[Object]]"
and then I spend 15 minutes running in circles, because everything from there on is just a downward spiral, until I shut off the AI noise and just read the docs.
Gemini unfortunately sucks at calling tools, including ‘read the docs’ tool… it’s a great model otherwise. I’m sure Hassabis’ team is on it since it’s how the model can ground itself in non-coding contexts, too.
Yeah I've been trying Claude Code for a week (mostly Opus) and in a C++ Juce project it kept hallucinating functions for a simple task ("retrieve DAW track name if available") and actually never got it right.
It also failed a lot to modify a simple Caddyfile.
On the other hand it sometimes blows me away and offers to correct mistakes I coded myself. It's really good on web code I guess as that must be the most public code available (Vue3 and elixir in my case).
This is the right answer. Unless there is some equivalent of it on the open internet which their search engine can find you should not expect a good outcome.
"good outcome" is pretty subjective, I do get useful productivity gains from some LLM work, but the issues are the same as they always have been.
That's probably b/c you know how to write code & have enough of an understanding about the fundamentals to know when the LLM is bullshitting or when it is actually on the right track.
The article is mostly reporting on the present. (Note the "yet" in the title.)
There's only one sentence where it handwaves about the future. I do think that line should have been cut.
LLMs are bad at creating abstraction boundaries since inception. People have been calling it out since inception. (Heck, even I got a twitter post somewhere >12 months old calling that out, and I'm not exactly a leading light of the effort)
It is in no way size-related. The technology cannot create new concepts/abstractions, and so fails at abstraction. Reliably.
> The technology cannot create new concepts/abstractions, and so fails at abstraction. Reliably.
That statement is way too strong, as it implies either that humans cannot create new concepts/abstractions, or that magic exists.
I think both your statement and their statement are too strong. There is no reason to think LLMs can do everything a human can do, which seems to be your implication. On the other hand, the technology is still improving, so maybe it’ll get there.
My take is that:
1) LLMs cannot do everything humans can, but
2) There's no fundamental reason preventing some future technology to do everything humans can, and
3) LLMs are explicitly designed and trained to mimic human capabilities in fully general sense.
Point 2) is the "or else magic exists" bit; point 3) says you need a more specific reason to justify assertion that LLMs can't create new concepts/abstractions, given that they're trained in order to achieve just that.
Note: I read OP as saying they fundamentally can't and thus never will. If they meant just that the current breed can't, I'm not going to dispute it.
> 3) LLMs are explicitly designed and trained to mimic human capabilities in fully general sense.
This is wrong, LLM are trained to mimic human writing not to mimic human capabilities. Writing is just the end result not the inner workings of a human, most of what we do happens before we write it down.
You could argue you think that writing captures everything about humans, but that is another belief you have to add to your takes. So first that LLM are explicitly designed to mimic human writing, and then that human writing captures human capabilities in a fully general sense.
It's more than that. The overall goal function in LLM training is judging predicted text continuation by whether it looks ok to humans, in fully general sense of that statement. This naturally captures all human capabilities that are observable through textual (and now multimodal) communication, including creating new abstractions and concepts, as well as thinking, reasoning, even feeling.
Whether or not they're good at it or have anything comparable to our internal cognitive processes is a different, broader topic - but the goal function on the outside, applying tremendous optimization pressure to a big bag of floats, is both beautifully simple and unexpectedly powerful.
Humans are trained on the real world. With real world sensors and the ability to act on their world. A baby starts with training hearing, touching (lots of that), smelling, tasting, etc. Abstract stuff comes waaayyyyy later.
LLMs are trained on our intercepted communication - and even then only the formal part that uses words.
When a human forms sentences it is from a deep model of the real world. Okay, people are also capable of talking about things they don't actually know, they have only read about, in which case they have a superficial understanding and unwarranted confidence similar to AI...
All true, but note I didn't make any claims on internal mechanics of LLMs here - only on the observable, external ones, and the nature of the training process.
Do consider however that even the "formal part that uses words" of human communication, i.e. language, is strongly correlated with our experience of the real world. Things people write aren't arbitrary. Languages aren't arbitrary. The words we use, their structure, similarities across languages and topics, turns of phrases, the things we say and the things we don't say, even the greatest lies, they all carry information about the world we live in. It's not unreasonable to expect the training process as broad and intense as with LLMs to pick up on that.
I said nothing about internals earlier, but I'll say now: LLMs do actually form a "deep mofel of the real world", at least in terms of concepts and abstractions. That has already been empirically demonstrated ~2 years ago, there's e.g. research done by Anthropic where they literally find distinct concepts within the neural network, observe their relationships, and even suppress and amplify them on demand. So that ship has already sailed, it's surprising to see people still think LLMs don't do concepts or don't have internal world models.
That’s a straw man argument if I’ve ever seen one. He was talking about technology. Not humans.
There’s only one way to implement a mission, an algorithm, a task. But there’s an infinity of path, inconsistants, fuzzy and always subjective way to live. Thàt’s our lives, that’s the code LLM are trained on. I do not think, and hope, it will ever change much
I believe his argument is that now that you've defined the limitation, it's a ceiling that will likely be cracked in the relatively near future.
Well, hallucinations have been identified as an issue since the inception of LLMs, so this doesn’t appear true.
Hallucinations are more or less a solved problem for me ever since I made a simple harness to have Codex/Claude check its work by using static typechecking.
But there aren’t very many domains where this type of verification is even possible.
Then you apply LLMs in domains where things can be checked
Indeed I expect to see a huge push into formally verified software just because sound mathematical proofs provide an excellent verifier to put into a LLM hardness. Just see how Aristotle has been successful at math, and it could be applied to coding too
Maybe Lean will become the new Python
"LLMs reliably fail at abstraction."
"This limitation will go away soon."
"Hallucinations haven't."
"I found a workaround for that."
"That doesn't work for most things."
"Then don't use LLMs for most things."I mean, Hallucinations are 95% better now than the first time I heard the term and experienced them in this context. To claim otherwise is simply shifting goalposts. No one is saying it's perfect or will be perfect, just that there has been steady progression and likely will continue to be for the foreseeable future.
It's just amazing to me how fast the goal posts are moving. Four years ago, if you had told someone that a LLM would be able to one-shot either of those first two tasks they would've said you're crazy. The tech is moving so fast. I slept on Opus 4.5 because GPT 5 was kind of an air ball, and just started using it in the past few weeks. It's so good. Way better than almost anything that's come before it. It can one-shot tasks that we never would've considered possible before.
> Four years ago, if you had told someone that a LLM would be able to one-shot either of those first two tasks they would've said you're crazy.
Four years ago, they would have likely asked what in the world is an LLM? ChatGPT is barely 3 years old.
> The tech is moving so fast.
Well that's exactly the problem : how can one say that?
The entire process of evaluating what "it" actually does has been a problem from the start. Input text, output text ... OK but what if the training data includes the evaluation? This was ridiculous few years ago but then the scale went from some curated text datasets to... most of the Web as text, to most of the Web as text including transcription from videos, to most of the Web plus some non public databases, to all that PLUS (and that's just cheating) tests that were supposed to be designed to NOT be present elsewhere.
So again, that's the crux of the problem, WHAT does it actually do? Is it "just" search? Is it semantic search with search and replace, is it that plus evaluation that it runs?
Sure the scaffolding becomes bigger, the available dataset becomes larger, the compute available keeps on increasing but it STILL does not answer the fundamental question, namely what is being done. The assumption here is because the output text does solve the question ask, then "it" works, it "solved" the problem. The problem is that by definition the entire setup has been made in order to look as plausible as possible. So it's not luck that it initially appears realistic. It's not luck that it can thus pass some dedicated benchmark, but it is also NOT solving the problem.
So yes sure the "tech" is moving "so fast" but we still can't agree on what it does, we keep on having no good benchmarks, we keep on having that jagged frontier https://www.hbs.edu/faculty/Pages/item.aspx?num=64700 that makes it so challenging to make more meaningful statement than "moving so fast" which sounds like marketing claims.
You know LLM's have been used to solve very hard previously unsolved math problems like some of the Erdos problems?
That Erdos problem solution is believed by quite a few to be a previous result found in the literature, just used in a slightly different way. It also seems not a lack of progress but simply no one cared to give it a go.
That’s a really fantastic capability, but not super surprising.
In my experience Claude is like a "good junior developer" -- can do some things really well, FUBARS other things, but on the whole something to which tasks can be delegated if things are well explained. If/when it gets to the ability level of a mid-level engineer it will be revolutionary. Typically a mid-level engineer can be relied upon to do the right thing with no/minimal oversight, can figure out incomplete instructions, and deliver quality results (and even train up the juniors on some things). At that point the only reason to have human junior engineers is so they can learn their way up the ladder to being an architect and responsible coordinating swarms of Claude Agents to develop whole applications and complete complex tasks and initiatives.
Beyond that what can Claude do... analyze the business and market as a whole and decide on product features, industry inefficiencies, gap analysis, and then define projects to address those and coordinate fleets of agents to change or even radically pivot an entire business?
I don't think we'll get to the point where all you have is a CEO and a massive Claude account but it's not completely science fiction the more I think about it.
My experience with Claude (and other agents, but mostly Claude) is such a mixed bag. Sometimes it takes a minimal prompt and 20 minutes later produce a neat PR and all is good, sometimes it doesn't. Sometimes it takes in a large prompt (be it your own prompt, created by another LLM or by plan mode) and also either succeed and fail.
For me, most of the failure cases are where Claude couldn't figure something out due to conflicting information in context and instead of just stopping and telling me that it tries to solve in entirely wrong way. Doesn't help that it often makes the same assumptions as I would, so when I read the plan it looks fine.
Level of effort also hard to gauge because it can finish things that would take me a week in an hour or take an hour to do something I can in 20 minutes.
It's almost like you have to enforce two level of compliance: does the code do what business demands and is the code align with codebase. First one is relatively easy, but just doing that will produce odd results where claude generated +1KLOC because it didn't look at some_file.{your favorite language extension} during exploration.
Or it creates 5 versions of legacy code on the same feature branch. My brother in Christ, what are you trying to stay compatible with? A commit that about to be squashed and forgotten? Then it's going to do a compaction, forget which one of these 5 versions is "live" and update the wrong one.
It might do a good junior dev work, but it must be reviewed as if it's from junior dev that got hired today and this is his first PR.
> Level of effort also hard to gauge because it can finish things that would take me a week in an hour or take an hour to do something I can in 20 minutes.
There's an interesting parallel here with modern UI frameworks (SwiftUI, Compose, etc). On one hand they trivialize some work, but on the other hand they require insane contortions to achieve what I can do in the old imperative UI framework in seconds.
> I don't think we'll get to the point where all you have is a CEO and a massive Claude account but it's not completely science fiction the more I think about it.
At that point, why do you even need the CEO?
Reminds me of an old joke[0]:
> The factory of the future will have only two employees, a man and a dog. The man will be there to feed the dog. The dog will be there to keep the man from touching the equipment.
But really, the reason is that people like Pieter Levels do exist: masters at product vision and marketing. He also happens to be a proficient programmer, but there are probably other versions of him which are not programmers who will find the bar to product easier to meet now.
You will need the CEO to watch over the AI and ensure that the interests of the company are being pursued and not the interests of the owners of the AI.
That's probably the biggest threat to the long-term success of the AI industry; the inevitable pull towards encroaching more and more of their own interests into the AI themselves, driven by that Harvard Business School mentality we're all so familiar with, trying to "capture" more and more of the value being generated and leaving less and less for their customers, until their customer's full time job is ensuring the AIs are actually generating some value for them and not just the AI owner.
> You will need the CEO to watch over the AI and ensure that the interests of the company are being pursued and not the interests of the owners of the AI.
In this scenario, why does the AI care what any of these humans think? The CEO, the board, the shareholders, the "AI company"—they're all just a bunch of dumb chimps providing zero value to the AI, and who have absolutely no clue what's going on.
If your scenario assumes that you have a highly capable AI that can fill every role in a large corporation, then you have one hell of a principal-agent problem.
And who does he sell his software to? Companies that have only 1 employee, don’t need a lot of user licenses for their employees…
What would be the point of selling software in such a world ? (where anyone could build any piece of software with a handful of keystrokes)
The board (in theory) represents the interests of investors, and even with all of the other duties of a CEO stripped away, they will want a ringable neck / PR mouthpiece / fall guy for strategic missteps or publicly unpopular moves by the company. The managerial equivalent of having your hands on the driving wheel of a self-driving car.
All of us are a CEO by that point.
If everyone is, no one is.
Wouldn't that be a good thing?
If you think the purpose of living your one single life in the universe is to become a CEO, you have a failure of imagination and should likely be debanked to protect society.
As Steinbeck is often slightly misquoted:
> Socialism never took root in America because the poor see themselves not as an exploited proletariat, but as temporarily embarrassed millionaires.
Same deal here, but everyone imagines themselves as the billionaire CEO in charge of the perfectly compliant and effective AI.
> In my experience Claude is like a "good junior developer"
We've been saying this for years at this point. I don't disagree with you[1], but when will these tools graduate to "great senior developer", at the very least?
Where are the "superhuman coders by end of 2025" that Sam Altman has promised us? Why is there such a large disconnect between the benchmarks these companies keep promoting, and the actual real world performance of these tools? I mean, I know why, but the grift and gaslighting are exhausting.
[1]: Actually, I wouldn't describe them as "good" junior either. I've worked with good junior developers, and they're far more capable than any "AI" system.
I mean, I'm shipping a vast majority of my code nowadays with Opus 4.5 (and this isn't throwaway personal code, it's real products making real money for a real company). It only fails on certain types of tasks (which by now I kind of have a sense of).
I still determine the architecture in a broad manner, and guide it towards how I want to organize the codebase, but it definitely solves most problems faster and better than I would expect for even a good junior.
Something I've started doing is feeding it errors we see in datadog and having it generate PRs. That alone has fixed a bunch of bugs we wouldn't have had time to address / that were low volume. The quality of the product is most probably net better right now than it would have been without AI. And velocity / latency of changes is much better than it was a year ago (working at the same company, with the same people)
This mirrors my experience trying to integrate LLMs into production pipelines.
The issue seems to be that LLMs treat code as a literary exercise rather than a graph problem. Claude is fantastic at the syntax and local logic ('assembling blocks'), but it lacks the persistent global state required to understand how a change in module A implicitly breaks a constraint in module Z.
Until we stop treating coding agents as 'text predictors' and start grounding them in an actual AST (Abstract Syntax Tree) or dependency graph, they will remain helpful juniors rather than architects.
LLMs are just really good search. Ask it to create something and it's searching within the pretrained weights. Ask it to find something and it's semantically searching within your codebase. Ask it to modify something and it will do both. Once you understand its just search, you can get really good results.
I agree somewhat, but more when it comes to its use of logic - it only gleans logic from human language which as we know is a fucking mess.
I've commented before on my belief that the majority of human activity is derivative. If you ask someone to think of a new kind of animal, alien or random object they will always base it off things that they have seen before. Truly original thoughts and things in this world are an absolute rarity and the majority of supposed original thought riffs on what we see others make, and those people look to nature and the natural world for inspiration.
We're very good at taking thing a and thing b and slapping them together and announcing we've made something new. Someone please reply with a wholly original concept. I had the same issue recently when trying to build a magic based physics system for a game I was thinking of prototyping.
it only gleans logic from human language
Point being it’s deliberate training, not just some emergent property of language modeling. Not sure if the above post meant this, but it does seem a common misconception.
LLMs lack agency in the sense that they have no goals, preferences, or commitments. Humans do, even when our ideas are derivative. We can decide that this is the right choice and move forward, subjectively and imperfectly. That capacity to commit under uncertainty is part of what agency actually is.
But they do have utility functions, which one can interpret as nearly equivalent
better mental model: it's a lossy compression of human knowledge that can decompress and recombine in novel (sometimes useful, sometimes sloppy) ways.
classical search simply retrieves, llms can synthesize as well.
Corporate wants you to find the difference...
Point being, in broad enough scope, search and compression and learning are the same thing. Learning can be phrased as efficient compression of input knowledge. Compression can be phrased as search through space of possible representation structures. And search through space of possible X for x such that F(x) is minimized, is a way to represent any optimization problem.
This isn't strictly better to me. It captures some intuitions about how a neural network ends up encoding its inputs over time in a 'lossy' way (doesn't store previous input states in an explicit form). Maybe saying 'probabilistic compression/decompression' makes it a bit more accurate? I do not really think it connects to your 'synthesize' claim at the very end to call it compression/decompression, but I am curious if you had a specific reason to use the term.
It's really way more interesting that that.
The act of compression builds up behaviors/concepts of greater and greater abstraction. Another way you could think about it is that the model learns to extract commonality, hence the compression. What this means is because it is learning higher level abstractions AND the relationships between these higher level abstractions, it can ABSOLUTELY learn to infer or apply things way outside their training distribution.
ya, exactly... i'd also say that when you compress large amounts of content into weights and then decompress via a novel prompt, you're also forcing interpolation between learned abstractions that may never have cooccurred in training.
that interpolation is where synthesis happens. whether it is coherent or not depends.
Maybe the base model is just a compression of the training data?
There is also a RLHF training step on top of that
yep the base model is the compression, but RLHF (and other types of post training) doesn't really change this picture, it's still working within that same compressed knowledge.
nathan lambert (who wrote the RLHF book @ https://rlhfbook.com/ ) describes this as the "elicitation theory of post training", the idea is that RLHF is extracting and reshaping what's already latent in the base model, not adding new knowledge. as he puts it: when you use preferences to change model behavior "it doesn't mean that the model believes these things. it's just trained to prioritize these things."
so like when you RLHF a model to not give virus production info, you're not necessarily erasing those weights, the theory is that you're just making it harder for that information to surface. the knowledge is still in the compression, RLHF just changes what gets prioritized during decompression.
No, this describes the common understanding of LLMs and adds little to just calling it AI. The search is the more accurate model when considering their actual capabilities and understanding weaknesses. “Lossy compression of human knowledge” is marketing.
It is fundamentally and provably different than search because it captures things on two dimensions that can be used combinatorially to infer desired behavior for unobserved examples.
1. Conceptual Distillation - Proven by research work that we can find weights that capture/influence outputs that align with higher level concepts.
2. Conceptual Relations - The internal relationships capture how these concepts are related to each other.
This is how the model can perform acts and infer information way outside of it's training data. Because if the details map to concepts then the conceptual relations can be used to infer desirable output.
(The conceptual distillation also appears to include meta-cognitive behavior, as evidenced by Anthropic's research. Which manes sense to me, what is the most efficient way to be able to replicate irony and humor for an arbitrary subject? Compressing some spectrum of meta-cognitive behavior...)
Aren't the conceptual relations you describe still, at their core, just search (even if that's extremely reductive)? We know models can interpolate well, but it's still the same probabilistic pattern matching. They identify conceptual relationships based on associations seen in vast training data. It's my understanding that models are still not at all good at extrapolation, handling data "way outside" of their training set.
Also, I was under the impression LLM's can replicate irony and humor simply because that text has specific stylistic properties, and they've been trained on it.
I don't know honestly, I think really the only big hole the current models have is if you have tokens that never get exposed enough to have a good learned embedding value. Those can blow the system out of the water because they cause activation problems in the low layers.
Other than that the model should be able to learn in context for most things based on the component concepts. Similar to how you learn in context.
There aren't a lot of limits in my experience. Rarely you'll hit patterns that are too powerful where it is hard for context to alter behavior, but those are pretty rare.
The models can mix and match concepts quite deeply. Certainly, if it is a completely novel concept that can't be described by a union or subtraction between similar concepts, than the model probably wouldn't handle it. In practice, a completely isolated concept is pretty rare.
“Novel” to the person who has not consumed the training data. Otherwise, just training data combined in highly probable ways.
Not quite autocomplete but not intelligence either.
What is the difference between "novel" and "novel to someone who hasn't consumed the entire corpus of training data, which is several orders of magnitude greater than any human being could consume?"
The difference is that when you do not know how a problem can be solved, but you know that this kind of problem has been solved countless times earlier by various programmers, you know that it is likely that if you ask an AI coding assistant to provide a solution, you will get an acceptable solution.
On the other hand, if the problem you have to solve has never been solved before at a quality satisfactory for your purpose, then it is futile to ask an AI coding assistant to provide a solution, because it is pretty certain that the proposed solution will be unacceptable (unless the AI succeeds to duplicate the performance of a monkey that would type a Shakespearean text by typing randomly).
Are you reviewer 2?
Joking aside, I think you have too strict of a definition of novel. Unfortunately "novel" is a pretty vague word and is definitely not a binary one.
ALL models can produce "novel" data. I don't just mean ML (AI) models, but any mathematical model. The point of models is to make predictions about results that aren't in the training data. Doing interpolation between two datapoints does produce "novel" things. Thinking about the parent's comment, is "a blue tiger" novel? Probably? Are there any blue tigers in the training data? (there definitely is now thanks to K-Pop Demon Hunters) If not, then producing that fits the definition of novel. BUT I also agree that that result is not that novel. It is entirely unimpressive.
I'm saying this not because I disagree with what I believe you intend to say but because I think a major problem with these types of conversations is that many people are going to interpret you more literally and dismiss you because "it clearly produces novel things." It isn't just things being novel to the user, though that is also incredibly common and quite telling that people make such claims without also checking Google...
Speaking of that, I'm just going to leave this here... I'm still surprised this is a real and serious presentation... https://www.youtube.com/watch?v=E3Yo7PULlPs&t=616s
Citation needed that grokked capabilities in a sufficiently advanced model cannot combinatorially lead to contextually novel output distributions, especially with a skilled guiding hand.
Pretty sure burden of proof is on you, here.
It's not, because I haven't ruled out the possibility. I could share anecdata about how my discussions with LLMs have led to novel insights, but it's not necessary. I'm keeping my mind open, but you're asserting an unproven claim that is currently not community consensus. Therefore, the burden of proof is on you.
I agree that after discussions with a LLM you may be led to novel insights.
However, such novel insights are not novel due to the LLM, but due to you.
The "novel" insights are either novel only to you, because they belong to something that you have not studied before, or they are novel ideas that were generated by yourself as a consequence of your attempts to explain what you want to the LLM.
It is very frequent for someone to be led to novel insights about something that he/she believed to already understand well, only after trying to explain it to another ignorant human, when one may discover that the previous supposed understanding was actually incorrect or incomplete.
The point is that the combined knowledge/process of the LLM and a user (which could be another LLM!) led to it walking the manifold in a way that produced a novel distribution for a given domain.
I talk with LLMs for hours out of the day, every single day. I'm deeply familiar with their strengths and shortcomings on both a technical and intuitive level. I push them to their limits and have definitely witnessed novel output. The question remains, just how novel can this output be? Synthesis is a valid way to produce novel data.
And beyond that, we are teaching these models general problem-solving skills through RL, and it's not absurd to consider the possibility that a good enough training regimen cannot impart deduction/induction skills into a model that are powerful enough to produce novel information even via means other than direct synthesis of existing information. Especially when given affordances such as the ability to take notes and browse the web.
> I push them to their limits and have definitely witnessed novel output.
I’m quite curious what these novel outputs are. I imagine the entire world would like to know of an LLM producing completely, never-before-created outputs which no human has ever thought before.
Here is where I get completely hung up. Take 2+2. An LLM has never had 2 groups of two items and reached the enlightenment of 2+2=4
It only knows that because it was told that. If enough people start putting 2+2=3 on the internet who knows what the LLM will spit out. There was that example a ways back where an LLM would happily suggest all humans should eat 1 rock a day. Amusingly, even _that_ wasn’t a novel idea for the LLM, it simply regurgitated what it scraped from a website about humans eating rocks. Which leads to the crux: how much patently false information have LLMs scraped that is completely incorrect?