by nkko on 5/9/25, 4:55 AM with 136 comments
by haberman on 5/9/25, 6:59 AM
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).
by sschueller on 5/9/25, 6:40 AM
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.
by stevage on 5/9/25, 8:42 AM
It feels like a hand-crafted algorithm would get a much better result.
by jader201 on 5/9/25, 6:06 AM
by RaSoJo on 5/9/25, 9:26 AM
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.
by yathaid on 5/9/25, 6:26 AM
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
by psiops on 5/9/25, 7:47 AM
by gilgoomesh on 5/9/25, 6:15 AM
by kilimounjaro on 5/9/25, 6:56 AM
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?
by carstenhag on 5/9/25, 6:34 AM
by W0lfEagle on 5/9/25, 6:35 AM
by dwighttk on 5/9/25, 11:05 AM
by 9dev on 5/9/25, 6:35 AM
by soared on 5/9/25, 1:14 PM
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.
by 6stringmerc on 5/9/25, 4:44 PM
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.
by oaiey on 5/9/25, 8:12 AM
by z3t4 on 5/9/25, 10:41 AM
by yunusabd on 5/9/25, 6:57 AM
by vladde on 5/9/25, 7:36 AM
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?
by necovek on 5/9/25, 12:05 PM
by andyjohnson0 on 5/9/25, 6:27 PM
by josefx on 5/9/25, 7:55 AM
by belter on 5/9/25, 11:32 AM
by Traubenfuchs on 5/9/25, 7:01 AM
Can it produce an ample bossom made of lego? And indecent protrusion? Weapons?
by sargstuff on 5/10/25, 6:46 PM
by londons_explore on 5/9/25, 8:40 AM
I guess I learned a word today...
by m3kw9 on 5/9/25, 3:03 PM
by xeyownt on 5/9/25, 7:19 AM
Sometimes the amount of money and energy that are spent in "recreation" projects just amazes me.
by benob on 5/9/25, 6:20 AM
by serial_dev on 5/9/25, 8:14 PM
by nurettin on 5/9/25, 6:47 AM