[HN Gopher] LegoGPT: Generating Physically Stable and Buildable ...
       ___________________________________________________________________
        
       LegoGPT: Generating Physically Stable and Buildable Lego
        
       Author : nkko
       Score  : 535 points
       Date   : 2025-05-09 04:55 UTC (18 hours ago)
        
 (HTM) web link (avalovelace1.github.io)
 (TXT) w3m dump (avalovelace1.github.io)
        
       | jader201 wrote:
       | There's a bug on the page (on iPhone, at least) once you scroll
       | to the gifs that it starts to auto load them without doing
       | anything, making it really hard to navigate anywhere at that
       | point.
        
         | Aeolun wrote:
         | When will people finally learn to _never_ autoplay.
        
           | vachina wrote:
           | Autoplay is fine, it's Safari opting to autoplay in
           | FULLSCREEN. Firefox et. al. play them in the respective video
           | containers.
        
         | pragmatick wrote:
         | The opposite for me on Firefox Desktop - I didn't realize they
         | were gifs and wondered what the pictures were supposed to tell
         | me.
        
         | MangoTec wrote:
         | this should be fixable with `playsinline` on the video element:
         | https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/...
         | 
         | annoying that this is the default behaviour on iOS though
        
       | gilgoomesh wrote:
       | It's hilarious watching $50,000 worth of robots take so long to
       | assemble a couple dollars worth of Lego. It's like peering into
       | the old folks home for robots.
        
         | FirmwareBurner wrote:
         | That should tell you why stuff is still hand assembled in Asia
         | instead of by robots in the west.
        
           | smikhanov wrote:
           | Also why it's OK to stop worrying about our future robotic
           | (or AI) overlords.
        
             | FirmwareBurner wrote:
             | Those are already an issue. AI is a bigger threat to
             | cognitive tasks than to physical ones.
             | 
             | Skynet isn't goanna attack you with Terminators wielding a
             | _" phased plasma rifle in the 40W range"_, but will be
             | auto-rejecting your job application, your health insurance
             | claims, your credit score and brain washing your relatives
             | on social media.
        
               | smikhanov wrote:
               | Absolutely, that's without any doubt.
               | 
               | There's a difference though. The "cool" Terminator Skynet
               | pursues its own goals, and wasn't programmed by humans to
               | kill. The "boring" insurance-rejecting Skynet is
               | explicitly programmed to reject insurance claims by other
               | humans, unfortunately.
               | 
               | So still, no need to worry about our AI overlords, worry
               | about people running the AI systems.
        
               | imtringued wrote:
               | > AI is a bigger threat to cognitive tasks than to
               | physical ones.
               | 
               | I don't see how you could possibly think this is true.
               | Physical automation is easier to scale since you only
               | need to solve a single problem instance and then just
               | keep applying it on a bigger scale.
        
               | FirmwareBurner wrote:
               | Automation doesn't work where high dexterity and quick
               | adaptability is required. You can much cheaper and
               | quicker to train a human worker to move from sewing a
               | Nike shoe to an Adidas shoe than you can reprogram and
               | retool a robot.
               | 
               | Robots work for highly predictable high speed tasks where
               | dexterity is not an issue, like PCB pick and place.
        
               | davidthewatson wrote:
               | This reply is so perfect I'm going to memorize it for
               | family and friends.
        
           | femto wrote:
           | As a counterexample, I offer a pick-and-place line in action.
           | 
           | https://youtu.be/Ca-SoKzjh4M?t=110
           | 
           | SMT component placement isn't that different to placing
           | bricks. Conventional wisdom is that if you can design a PCB
           | that requires no manual work, its assembly cost is more-or-
           | less location independent. SMT pick and place can hit speeds
           | of 200,000 components per hour [1]. That's about 50
           | components per second.
           | 
           | [1] https://www.hallmarknameplate.com/smt-process/
        
             | imtringued wrote:
             | Fixturing isn't automated in most places. Sure a gantry
             | style CNC machine can drive screws vertically into your
             | parts to join them, but it requires a human loader to put
             | the two parts onto the fixture in the first place.
        
             | FirmwareBurner wrote:
             | The tasks requiring high dexterity like final assembly of
             | the product with displays, keyboards, ribbon cables and
             | cases is still done by humans by hand.
        
         | Zobat wrote:
         | People claim that Lego is expensive, but try buying a robot
         | that builds Lego...
        
           | bombcar wrote:
           | You build the robot out of Lego.
        
             | wiz21c wrote:
             | They should have done it with lego mindstorm :-)
        
               | bombcar wrote:
               | Someone has. I can't find it, but there was a Lego robot
               | that would build a (simple) Lego set.
        
         | KurSix wrote:
         | Give it a decade and we'll probably have robo-builders doing it
         | faster than we can blink...
        
           | cruffle_duffle wrote:
           | I would certainly hope the laundry robots come first. Screw
           | Lego robots and self driving cars. Please just take the
           | laundry out of the dryer, fold it all and put it away.
        
           | LargeWu wrote:
           | First AI is creating our art and pretty soon it will be
           | playing with our toys for us too
        
       | benob wrote:
       | It looks like it is an extension of 3d model generation
       | techniques.
        
       | yathaid wrote:
       | This is super cool! The GIFs showing the object being built are
       | just yummy; I have no other way to describe it.
       | 
       | If anyone else was searching for the dataset, it is at
       | https://huggingface.co/datasets/AvaLovelace/StableText2Lego
       | 
       | It contains " contains 47,000+ different LEGO structures,
       | covering 28,000+ unique 3D objects from 21 common object
       | categories of the ShapeNetCore dataset".
       | 
       | Local inference instructions are over at their github page -
       | https://github.com/AvaLovelace1/LegoGPT/?tab=readme-ov-file
        
       | carstenhag wrote:
       | Have the authors never heard of Lego being one of the companies
       | that are super strict about their trademark? They file takedown
       | notices etc on every project they see. Even if the stone design
       | has the little thingies on top/bottom...
        
       | W0lfEagle wrote:
       | Great. Please do cabinets next. Constrain to some specified
       | material such as 2.5m by 1.25m 18mm ply. Iterate designs by text
       | and output the model, cutlist and assembly instructions. Simple
       | right?
        
       | 9dev wrote:
       | When I was a kid, I proudly exclaimed I wanted to become a
       | professional lego builder. Not in my wildest dreams would I have
       | assumed how close to that career path I could have come.
        
       | sschueller wrote:
       | This is probably going to get a letter from LEGO's lawyers.
       | 
       | If you want to be safe do not use the word LEGO. Use Bricks or in
       | German "Klemmbausteine".
       | 
       | Many people have had to deal with LEGO's lawyers and it ain't
       | pretty.
        
         | edoceo wrote:
         | Why are they like Nintendo when they could be like Sega?
         | Embrace your community where they are.
        
           | raverbashing wrote:
           | (Not saying it's related) But, which one of those are still
           | running?
        
             | 71bw wrote:
             | Both?
        
             | vanderZwan wrote:
             | Both are. Sega just lost a console war a few decades ago
             | and decided not to pursue that any more.
        
             | Cthulhu_ wrote:
             | Sega is generating more than $1.5 billion a year, they're
             | fine.
        
             | Philpax wrote:
             | I'd go as far as to say that Sega's embracing of their fans
             | is a big part of why they're still around:
             | https://en.wikipedia.org/wiki/Sonic_Mania#Development
        
           | Freak_NL wrote:
           | Trademark law leaves no space for that. The Lego Group has to
           | _actively_ defend their trademark. That means a name like
           | LegoGPT is really on the obvious end of  'don't do that'.
        
             | makeitdouble wrote:
             | https://www.eff.org/deeplinks/2013/11/trademark-law-does-
             | not...
        
               | shakna wrote:
               | Whilst they don't need to attack every form of speech, a
               | name like 'LegoGPT' is not protected.
        
               | dudeinjapan wrote:
               | If LEGO's lawyers agreed with this article, they'd be out
               | of business!
        
             | MrOrelliOReilly wrote:
             | Completely agree. This should be well beyond accusations of
             | corporate bullying. It's one thing to mention Legos, it's
             | another to actively include a brand name in your product!
             | NikeGPT, CocaColaGPT and IkeaGPT will face the same issue
             | ;)
        
               | Freak_NL wrote:
               | Mentioning Lego is absolutely OK, and you can sell used
               | Lego as well and note that you are using genuine Lego
               | bricks (resale laws simply allow that). Lego is really
               | antsy about anything which might look like it is actually
               | a Lego Group initiative though, and anything where Lego
               | bricks are offered for sale in a modified state1.
               | 
               | 1: Never, ever, sell _modified_ Lego bricks:
               | https://www.brickfanatics.com/lego-wins-court-case-
               | against-c...
        
             | cluckindan wrote:
             | The registered trademark is LEGO, in all caps.
             | 
             | Also, they don't tend to go after fan-made things like
             | this, based on some googling they typically throw the book
             | at counterfeit producers who are eating into their profits.
        
               | Freak_NL wrote:
               | (Initially) fan-made stuff which gets big enough to get
               | noticed usually won't be able to call themselves
               | something with 'Lego' in it. Usually some variation of
               | 'brick' is used instead (e.g., Bricklink, Rebrickable,
               | EuroBricks, etc.).
        
               | cluckindan wrote:
               | BrickLink is commercial, so is Rebrickable. EuroBricks
               | seems to have a user base which includes LEGO insiders.
        
               | 6stringmerc wrote:
               | Backing by a multimillion dollar University for
               | publication and promotional purposes is a far cry from
               | hobbyist and enthusiast ecosystems, at least in my view.
               | Then again I'm in a very slim minority of actual creative
               | creators who generate IP from scratch and my perspective
               | is much different than "move fast break things"
               | attitudes.
        
           | Perz1val wrote:
           | > Embrace your community where they are.
           | 
           | In the casinos?
        
           | makeitdouble wrote:
           | Sega's [0] main business is pachinko (so gambling). To them
           | Sonic brand being used by fans has very little consequences,
           | if not building most needed goodwill toward their other
           | brands.
           | 
           | That's where Nintendo is fundamentaly different.
           | 
           | [0] https://en.m.wikipedia.org/wiki/Sega_Sammy_Holdings
        
             | MrsPeaches wrote:
             | Interesting that many of Sega's games are now mobile
             | focused.
             | 
             | Also don't forget that Sega was "originally an importer of
             | coin-operated arcade games to Japan and manufacturer of
             | slot machines and jukeboxes"
             | 
             | https://en.m.wikipedia.org/wiki/History_of_Sega
        
             | foobahhhhh wrote:
             | Did that start with that merger in 2004 so back in the
             | Sonic heyday it wasn't in to gambling?
        
               | makeitdouble wrote:
               | Yes.
               | 
               | Sega was mostly into normal arcade games, and Sammy
               | baught them for their expertise to improve Sammy's much
               | more profitable gambling machines. It's Sammy's CEO that
               | took the lead, and Sonic and console games became a mere
               | side business.
        
             | andrewchilds wrote:
             | TIL the pachinko connection perfectly explains the
             | visual/sound/game design of the Sonic games.
        
           | ygouzerh wrote:
           | They probably have a culture of "patents".
           | 
           | They just won the market because historically they reused
           | existing locking bricks concept from a company called
           | Kiddicraft, found a way to make it more lockable... and
           | patent it before the original company and other companies
           | could implement it.
           | 
           | We can say that they became famous half fir engineering
           | reason, and half from their legal department...
        
         | ChrisRob wrote:
         | Immediately thought the same thing! This will get busted very
         | soon
        
         | amelius wrote:
         | This is academic research, and I suppose it falls under fair
         | use.
        
           | msiebuhr wrote:
           | IANAL, but EU law doesn't have "fair use". It does have a
           | _very specific_ set of uses where you don't have to ask for
           | permission (or pay). As I understand, it is more limited than
           | the US' "fair use" doctrine.
           | 
           | EU being EU, I can only imagine there's a bunch of particular
           | rules around research that may or may not work in the
           | authors' favor.
        
           | 6stringmerc wrote:
           | Fair Use is Copyright and has a four factor test. Trademark
           | is different than Copyright. Perhaps learning the difference
           | might be educational...and fiscally prudent.
        
             | amelius wrote:
             | Fair Use != Fair use
             | 
             | https://www.inta.org/fact-sheets/fair-use-of-trademarks-
             | inte...
        
         | KurSix wrote:
         | Even YouTubers and small hobby sites have gotten takedown
         | notices just for using the name in the wrong context
        
         | necovek wrote:
         | They are actively using actual LEGO bricks, and as such they
         | are not misrepresenting anything.
         | 
         | Where there is gray area is in them not clearly stating they
         | are unaffiliated with LEGO the company.
         | 
         | OTOH, they also don't seem to be looking to monetize anything,
         | so they are at lower risk from LEGO having a plausible claim
         | that they are hurting their sales.
        
           | dec0dedab0de wrote:
           | While it is perfectly valid to describe what they made as a
           | designer or builder for LEGO, I do not believe they are
           | allowed to use part of a trademark in a way that could be
           | trademarkable itself, so basically good for everything but
           | the name.
           | 
           | But then again IANAL, and that is just how I understand the
           | American law, and every country is different.
        
           | 6stringmerc wrote:
           | This is an incredibly ignorant perspective on the nature and
           | intent of trademark law and I'm hopeful you will learn about
           | reality one way or another. As the saying goes, your feelings
           | don't matter in court.
        
         | andrewchilds wrote:
         | Meanwhile, Amazon is perfectly happy to sell/promote this "Boy
         | and Tiger Adventure Blocks Set" which simultaneously rips off
         | LEGO and Calvin & Hobbes: https://www.amazon.com/Adventure-
         | Rotatable-Compatible-Charac...
        
           | Retric wrote:
           | Ripping off Calvin and hobs sure but not Lego.
           | 
           | https://en.wikipedia.org/wiki/Hilary_Page#Kiddicraft
           | 
           | LEGO is based off of earlier designs of interlocking bricks,
           | they are well known because they got really good at
           | affordability, high tolerances, and durability not because
           | they invented the concept. Beyond that the original
           | functional patents have long expired.
        
       | nurettin wrote:
       | This could prove useful for minecraft bot scene
        
         | davidthewatson wrote:
         | Indeed. I thought blocks world stuff would be amazing for early
         | childhood education. I'm guessing some labs are already there
         | since minecraft supports user-programmable models for years
         | though I dunno the details. I'd be happy to learn if anybody
         | knows of their evolution since the rise of AI.
        
       | kilimounjaro wrote:
       | Doesn't seem to add much to just converting a 3d model into
       | voxels and therefore bricks.
       | 
       | Using bricks other than 2x2 and 2x4 blocks creatively to make
       | interesting things is really important, i'm not sure what type if
       | algorithm would best auto generate beautiful MOCs however? Was
       | thinking of doing a $50000 kaggle comp for this, what do others
       | think?
        
       | yunusabd wrote:
       | Cool project, but judging from the videos, it looks like some of
       | them can't actually be built using those instructions. E.g. "A
       | backless bench with armrest" would require some bricks to float
       | in the air with no support while you're assembling the rest.
        
         | biofox wrote:
         | The design is sound, just not the order of assembly shown. For
         | the bench, the lower bricks are suspended by the upper ones, so
         | would need to be assembled separately before connecting to the
         | legs.
        
           | yunusabd wrote:
           | True, easy for a human, not so easy for a robot to go through
           | those extra steps. I wonder if they made it work with the
           | robots, because in the video they only show the robots
           | building from the bottom up.
        
           | Etherlord87 wrote:
           | Still, I think you wouldn't find such a design in actual Lego
           | set, because such floating bricks make the construction weak.
        
       | haberman wrote:
       | > To improve the stability of the resulting designs, we employ an
       | efficient validity check and physics-aware rollback during
       | autoregressive inference, which prunes infeasible token
       | predictions using physics laws and assembly constraints.
       | 
       | I'm far from an AI expert, but I've long felt that this is one of
       | the most interesting ways to use AI: to generate and optimize
       | possibilities _within_ a set of domain-specific constraints that
       | are programmed manually.
       | 
       | For example, imagine an AI that is designed to optimize traffic
       | light patterns. You want a hard constraint that no intersection
       | gives a combination of green lights that could cause collisions.
       | But within that set of constraints, which you could manually
       | specify, the AI could go wild trying whatever ideas it can come
       | up with.
       | 
       | At that point, the interesting work is deciding how to design the
       | problem space and the set of constraints. In this case it's a set
       | of lego bricks and how they can be built (and be stable).
        
         | Narew wrote:
         | I haven't read how they apply the constraint. But there is
         | similar stuff when you force llm to generate structured output
         | like Json format. llama.cpp allow to match a custom grammar for
         | example.
        
         | benterix wrote:
         | > to generate and optimize possibilities within a set of
         | domain-specific constraints
         | 
         | Well, yes, we've been doing this for several decades, many
         | people call it metaheuristics. There is a wide array of
         | algorithms in there. An excellent and light intro can be found
         | here: https://cs.gmu.edu/~sean/book/metaheuristics/
        
           | eurekin wrote:
           | Metaheurestics? I always thought it's similar to "I don't
           | know how many neurons to put in the hidden layer... and I
           | also don't know how many hidden layers I need, so, let's make
           | it a part of the optimisation problem to find out on it's
           | own".
        
             | PeterStuer wrote:
             | That is usually called Hyperparameter tuning.
        
             | benterix wrote:
             | As for hyperparameter tuning, the existing solutions such
             | as Optuna or Katib (in KubeFlow) also use metaheuristics,
             | e.g. CMA-ES.
        
           | jllyhill wrote:
           | Thanks, but some strange coincidence this is exactly the book
           | I have right now. In the introduction the author says, "I
           | think these notes would best serve as a complement to a
           | textbook". Do you happen to know any good textbooks on that
           | topic?
        
             | benterix wrote:
             | Everybody has their own preferences, what worked for me was
             | Metaheuristics: From Design to Implementation by this guy:
             | 
             | https://www.youtube.com/watch?v=ksK-XzkSQlk
        
           | mzl wrote:
           | Or more generally the whole field of combinatorial
           | optimization, of which metaheuristics is a (small) part.
        
             | dvfjsdhgfv wrote:
             | I believe you are right in principle, regarding the small
             | part. However my personal impression is that in practical
             | applications metaheuristics is huge (although these things
             | are hard to quantify).
        
           | haberman wrote:
           | The description in your link says:
           | 
           | > What is a Metaheuristic? A common but unfortunate name for
           | any stochastic optimization algorithm intended to be the last
           | resort before giving up and using random or brute-force
           | search. Such algorithms are used for problems where you don't
           | know how to find a good solution, but if shown a candidate
           | solution, you can give it a grade.
           | 
           | That sounds like "the AI came up with a solution where cars
           | can crash, let's give that solution a bad grade."
           | 
           | I was hoping for something more like "the problem is
           | specified such that invalid solutions aren't even
           | representable, so only acceptable solutions are considered."
        
             | dvfjsdhgfv wrote:
             | > I was hoping for something more like "the problem is
             | specified such that invalid solutions aren't even
             | representable, so only acceptable solutions are
             | considered."
             | 
             | It my (roughly) work this way. For example, when you do
             | hyperparameter tuning, you specify upper and lower bounds
             | (so that "invalid solutions aren't even representable").
             | 
             | The problem is, you often have no idea what will work and
             | what not, and e.g. your HPO algorithm might hit the bounds,
             | suggegsting that it might make sense to extend them before
             | the next run.
        
             | kmacdough wrote:
             | > I was hoping for something more like "the problem is
             | specified such that invalid solutions aren't even
             | representable, so only acceptable solutions are
             | considered."
             | 
             | How on earth would one come up with a model where "crashing
             | cars isnt't representable"? I don't think you recognize how
             | ill-defined and nonsensical this expectation is. Especially
             | when you consider that a such a car may encounter a
             | situation where a crash is unavoidable, where there's
             | certainly room for damage control. Sliding scales ALWAYS
             | work better for optimizations anyways, since regression is
             | so powerful.
        
               | haberman wrote:
               | I was speaking in the context of my original post, which
               | was specifically talking about traffic lights at
               | intersections and what combination of lights are allowed
               | to be green at the same time.
               | 
               | I think it would be fairly straightforward to enumerate,
               | given a set of lights at an intersection, which
               | combination of lights can be green without allowing cars
               | to cross paths. In other words, we're ruling out
               | combinations that are fundamentally unacceptable and
               | would never be seen in the real world (like "all lights
               | are green at the same time").
               | 
               | That gives the AI a set of acceptable combinations that
               | can be considered. Essentially the AI is choosing an
               | integer in the range 1-max for each intersection at each
               | point in time.
               | 
               | This doesn't eliminate the possibility of car crashes if
               | someone runs a red light. But it lets us constrain the
               | optimization problem to the set of green light
               | configurations that are actually feasible to deploy.
        
         | zelos wrote:
         | You'd probably use some kind of MILP or CLP based model for
         | that kind of thing, wouldn't you? The constraints define the
         | search space and the solver algorithm then explores it.
        
         | londons_explore wrote:
         | Fun thing to try:
         | 
         | Ask an LLM: "Say the word APPLE", but modify the code so the
         | logits of the token for Apple/apple/APPLE is permanently set to
         | -Inf - ie. the model cannot say that word.
         | 
         | The output ends up like this:
         | 
         | "Banana. Oh, just kidding. Banana. Oh, it's so tasty I said it
         | wrong. Lets try again: Orange. Whoops, I meant to say grape. No
         | I meant to say the tasty crunchy fruit known as a carrot".....
        
           | londons_explore wrote:
           | Note that OP's traffic light problem would suffer the same
           | problem.
           | 
           | Ie. a smart model, knowing it cannot say a word, will give
           | the next best solution - for example maybe saying "A P P L E"
           | or maybe "I'm afraid I'm not able to do that".
           | 
           | However, a constrained model does not know or understand its
           | own constraints, so keeps trying to do things which aren't
           | allowed - and even goes back and tries to redo these things
           | which aren't allowed, because to the model it is a mistake
           | which needs correcting.
        
             | adammarples wrote:
             | There's a whole field of solving constrainted optimization
             | and it doesn't really work like that, but they don't use
             | LLMs.
        
           | jcims wrote:
           | Like your brain when you know you know a word but it's just
           | not surfacing in your mind.
           | 
           | I'm guessing I'm not that different from the average human
           | and I can 'feel' something physically while I'm searching for
           | the word. I've always wondered what that was.
        
           | stavros wrote:
           | I saw this exact thing in a question about who was the first
           | composer, the model kept outputting Boethius and then saying
           | "NO!", as if it couldn't escape its own Freudian slips.
        
         | KurSix wrote:
         | Totally agree, this is where AI shines the most for me too. Let
         | humans define the rules of the game (like physics or traffic
         | safety), and let the AI explore the massive search space for
         | optimized solutions.
        
         | jgalt212 wrote:
         | like Combinatorial Chemistry, but we should probably just call
         | it AI Chemistry for the likes.
         | 
         | https://en.wikipedia.org/wiki/Combinatorial_chemistry
        
           | dvfjsdhgfv wrote:
           | Not just for the likes, for money. It looks like whatever
           | smart algorithms you use, if you slap "AI" on it, you're more
           | likely to get investment (if that's what you're after).
        
         | bob1029 wrote:
         | Error feedback seems to be the one thing that can unlock some
         | of the original promises.
         | 
         | For example, if you give a text-to-SQL bot access to the same
         | idea (e.g., error feedback from the SQL provider), it is much
         | more likely to succeed in generating valuable queries.
        
         | lgiordano_notte wrote:
         | Agree with this. Constraining generation with physics,
         | legality, or even tooling limits turns the model into a search-
         | and-validate engine instead of a word predictor. Closer to
         | program synthesis.
         | 
         | The real value is upstream: defining a problem space so well
         | that the model is boxed into generating something usable.
        
         | lolinder wrote:
         | A simple version of this that already shines with existing LLMs
         | is JSON Schema mode. You can go quite a long way towards making
         | illegal states unrepresentable and then turn a model loose in
         | the constrained sandbox, with the guarantee that anything it
         | produces will be at least valid if not correct: it's basically
         | type safety for LLM output.
         | 
         | The same mechanism that underlies JSON Schema support can be
         | applied to any sort of validation and correction, and yeah, I'd
         | love to see more of this kind of thing!
        
         | smokel wrote:
         | You might be interested in Reinforcement Learning [1]. By
         | giving the system a negative reward, it may eventually start
         | complying with safety rules. Still a good idea to keep the
         | harness in place during production use, though.
         | 
         | [1] https://en.wikipedia.org/wiki/Reinforcement_learning
        
       | Traubenfuchs wrote:
       | Is this aligned, safeguarded, censored?
       | 
       | Can it produce an ample bossom made of lego? And indecent
       | protrusion? Weapons?
        
         | b3lvedere wrote:
         | Weapons has been kinda done earlier: https://nostarch.com/flego
         | https://nostarch.com/legoguns.htm
        
       | xeyownt wrote:
       | So, besides training a LLM to generate build instructions for
       | lego model, they have robots to assemble these models, and they
       | applied 3D texture on 3D generated model (what for?).
       | 
       | Sometimes the amount of money and energy that are spent in
       | "recreation" projects just amazes me.
        
         | thenaturalist wrote:
         | You do realize that a system like LEGO is just an extremely
         | efficient and cheap proof of concept with a proxy material
         | (LEGO) for later real life applications of building X from
         | standardized components Y right?
         | 
         | This is interesting and seemingly quite applicable base
         | research and we move forward by being curious.
        
       | vladde wrote:
       | Looks amazing, with the arms building the piece!
       | 
       | However, the model "A high-backed chair" has some floating pieces
       | in the middle of the seat, that are fastened from above. Can
       | these robots handle building these?
        
       | psiops wrote:
       | I noticed that "a basic sofa" involves some placing some floating
       | bricks if built in the order of the animation. It hints at the
       | way this model generates the designs. The automated assembly of
       | generated LEGO structures using robots would have serious trouble
       | creating these designs I reckon.
        
         | sdoering wrote:
         | I came here to say that. I immediately thought: Wow, this works
         | in the assembled version, but not the way the assembly is being
         | animated. You would need to first build the base sofa layer
         | from two levels so that the upper layer keeps the lower layer
         | bricks in place. Only afterwards could it be put onto the legs.
        
           | paulluuk wrote:
           | Indeed, I would be very curious to see how their robots would
           | actually build that sofa. Although the robots aren't really
           | part of the model of course, they're just a little extra.
        
       | josefx wrote:
       | Does it have a sense of scale? Can it make a matching bookcase,
       | table and chair or are they all going to end up with a random
       | size?
        
       | oaiey wrote:
       | Real challenge are not LEGO System pieces but LEGO Technic pieces
       | where you do not have to build layers bottom-up but everything
       | inside out
        
       | londons_explore wrote:
       | > voxelizing it onto a grid and applying _legolization_
       | 
       | I guess I learned a word today...
        
       | stevage wrote:
       | This does not seem like a very impressive result. It's using such
       | a small set of bricks and the results don't really look much like
       | the intended thing.
       | 
       | It feels like a hand-crafted algorithm would get a much better
       | result.
        
         | otabdeveloper4 wrote:
         | What we need is an AI where you feed it some photos of your
         | pile of bricks and it generates you instructions based on the
         | bricks you have.
         | 
         | (Totally feasible with today's technology, but you'll need to
         | train your own specialized models.)
        
           | stevage wrote:
           | https://au.lifehacker.com/toys-board-
           | games/50948/news/scan-y...
        
           | amelius wrote:
           | What I'd be interested in most is a robot that can assemble a
           | model from a pile of bricks/parts.
        
           | dspillett wrote:
           | There already exists an app that will, from photos of your
           | pile, pick out models you can make from a large library of
           | existing models. Though IIRC that has been around long enough
           | that it isn't quite using what people are currently calling
           | AI (instead using older ML techniques for brick
           | identification, and a basic DB search to pick out the valid
           | plans for the resulting list of bricks).
        
           | Spellman wrote:
           | Already exists, put out by LEGO https://brickit.app/
        
             | Loughla wrote:
             | My experience with that is that it gives you unbelievably
             | simplistic builds, not the complicated things you might be
             | imagining.
        
         | KurSix wrote:
         | But I think the cool part here isn't photorealism, it's the
         | combo of language understanding and physical buildability
        
         | nkko wrote:
         | this is very cool considering it is a fine-tuned 1B model
        
         | tokai wrote:
         | The (fake) texturing is the only thing making it somewhat work.
         | As normal colored bricks it would just be lumps of lego.
        
       | RaSoJo wrote:
       | I don't need automation to build LEGO sets -- that's the fun
       | part, and I want to do it myself. What I need is automation
       | _after_ the build: to clean up, sort the bricks by color and
       | shape, and store them properly.
       | 
       | I just wish scientists would start by solving problems that
       | actually exist in the real world. There's real value -- and real
       | money -- in that.
        
         | KurSix wrote:
         | You're totally right: sometimes the real innovation isn't in
         | making the fun parts easier, it's in making the boring parts
         | disappear
        
         | lee-rhapsody wrote:
         | The issue with solving real-world problems is that it distracts
         | from publishing, which is all scientists are taught to care
         | about.
        
         | joeyparsons wrote:
         | came here to write this comment. clean up and sorting is the
         | top problem in lego land.
        
         | shagie wrote:
         | Sorting 2 Metric Tons of Lego
         | https://jacquesmattheij.com/sorting-two-metric-tons-of-lego/
         | (2017) https://news.ycombinator.com/item?id=14226889
         | 
         | World's first AI-powered LEGO sorting machine built with 10,000
         | LEGO bricks https://www.brothers-brick.com/2019/12/06/this-fan-
         | created-t... (2019)
         | https://news.ycombinator.com/item?id=21741834
        
       | z3t4 wrote:
       | The results are a bit underwhelming, considering what I'm used to
       | see in image generation, world generation in games, etc.
        
       | dwighttk wrote:
       | Quit trying to read the article after the 15th video went to full
       | screen and had to be dismissed hitting the tiny x in the upper
       | left... 3 more interfered with me trying to go back to this page
        
         | beklein wrote:
         | You can also grab the paper as a PDF here:
         | https://arxiv.org/pdf/2505.05469.
         | 
         | Keep in mind that these sites are run by AI researchers, not
         | dedicated UX teams at major tech companies--so the interface
         | can feel a bit rough around the edges. That said, your critique
         | is still valid; it's just fair to cut them a little slack given
         | their priorities.
        
       | belter wrote:
       | Can I have IKEA-GPT ? This weekend if possible?
        
         | davidthewatson wrote:
         | Indeed. I'm guessing legal is the only reason we don't have
         | 3d-printed ikea. Raymond Loewy FTW. But then, we'd have garages
         | full of bespoke n-of-1 junk instead of mass-made LLM (liminal-
         | labor-made) junk.
        
       | necovek wrote:
       | I love this: it'd be great if it also worked to give age-
       | appropriate designs as well (eg. 5 yo has less patience and
       | ability than an 8 yo).
        
       | aayushmaan45 wrote:
       | good
        
       | aayushmaan45 wrote:
       | wasd
        
       | soared wrote:
       | The high backed chair gif example is interesting - the way it's
       | animated it would completely fall apart and be unstable. But if
       | you built it in reverse, it would work fine.
       | 
       | But it also shows the weirdness of the solution - in places where
       | larger bricks make sense, multiple smaller bricks are used
       | instead. In a section where a 2x6 should be repeated, in on
       | instance of the repetition it uses tow 1x6s. It's weird.
       | 
       | Cool idea.
        
       | m3kw9 wrote:
       | These legos doesn't look very intricate even by 200 pc box
       | standards.
        
       | 6stringmerc wrote:
       | Is this use of the trademark approved by the LEGO owners by any
       | chance? I skimmed it and there didn't seem to be an indication
       | this has been endorsed by the company.
       | 
       | It kind of makes me suspicious of the integrity of Carnegie
       | Mellon if they will allow trademark infringement of this type
       | because, well, it does make me feel like I can shit in a bag and
       | call it a Carnegie Mellon Socking Stuffer without consequence.
        
       | andyjohnson0 wrote:
       | Its been a long day and and I can think of nothing to add other
       | than that those blue robot arms are both calming and visually
       | pleasant.
        
       | serial_dev wrote:
       | It would make sense for a car to have proper LEGO wheels.
        
       ___________________________________________________________________
       (page generated 2025-05-09 23:01 UTC)