Rendered at 19:40:29 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
kickingkeys 58 minutes ago [-]
Thank you for sharing this! did not expect to find my work on HN randomly haha
radarsat1 9 hours ago [-]
Beautiful.
I am reminded of a paper I was inspired by a long time ago [0], (okay it's 2018 so I guess just 8 years ago, but it feels like longer, from the before-times), that demonstrated learning brush strokes. At the time there was already a lot of work on GANs, but these are pixel-based methods, and I was really interested in the idea of how to derive descriptive methods of scene generation/understanding. I found this work really interesting because it combined RL and GAN techniques in a creative way. I miss that kind of research.
Now of course VLMs have shown that you can mix modalities in generalized sequence-to-sequence problems and it doesn't surprise me that this kind of thing is possible, but it's so nice to see it done well using modern techniques.
Really awesome! Been thinking about how to get LLMs to do generative art (yes, the pre-AI definition of generative art). Love to see this approach and results!
Brendinooo 4 hours ago [-]
This is really nice, and it's a good reminder that we're so early with regards to how AI can be used to make art.
And the fact that it's flowers creates a pretty nice mental model for it! One can plant seeds and cultivate the plants that grow from them but ultimately aren't in complete control of the outcome.
mskogly 13 hours ago [-]
Really liked your video presentation, thanks for sharing, especially the part of image generators locking us into a certain context immediately, reducing us to spectators instead of creatives.
Played a lot with p5js some years ago, might pick it up again and try some of your ideas. The reinforcement learning part sounds about above my skill level though. :)
kickingkeys 59 minutes ago [-]
Hi, this post wasn't made by me but thank you for the kind works :)
_boffin_ 2 days ago [-]
I think this actually might be one of the best ways to train people to use AI. I can see this honing people's prompting abilities and expressiveness, along with constraints and desired outcome.
Wild the possibilities
iambenm 6 hours ago [-]
It's Genetic Programming via LLM, cool!
retinaros 1 hours ago [-]
I did that on SVG mostly to teach it to draw pelicans but also to generalize it. most of the behavior is from SFT on the base model tho. RL is very ineficient at style or at least at generating novelty out of distrib.
mysterydip 2 days ago [-]
Makes me wonder, can any LLMs code in Logo? Could result in some interesting designs.
bombastic311 15 hours ago [-]
This is really really awesome
thrance 8 hours ago [-]
I've been building something similar, but for voxels. It's able to make pretty good models from just Python code (calling into a custom native module written in Rust). Better than I hoped it would, in fact, but it's still not perfect.
accomplishdent 2 days ago [-]
What is the JavaScript doing?
simonw 8 hours ago [-]
Drawing things with p5.js.
ACCount37 10 hours ago [-]
I wonder how the image generation models that generate SVGs work.
Are they trained roughly like this? Or is it an LLM conditioned on image? Or on diffusion latents from a model trained to emit SVG-compatible imagery?
13639366668 6 hours ago [-]
[flagged]
behnamoh 15 hours ago [-]
[flagged]
aflinik 11 hours ago [-]
Can you post some links to other similar projects you've seen? I'd love to compare different approaches
It seems pretty novel so I'm guessing they only read the title?
People have done plenty with SVGs but it's rare to see human-in-the-loop approaches
smokel 11 hours ago [-]
Interesting, tell me more!
pelasaco 13 hours ago [-]
March 2026 · Research, Design, Development.
Alien1Being 3 hours ago [-]
AI HYPE
yoloakki 5 hours ago [-]
For anyone looking to do this on their own, Datapoint (trydatapoint.com) lets you collect thousands of pairwise human preferences within minutes, and also has a pretty generous data grant for academics.
I am reminded of a paper I was inspired by a long time ago [0], (okay it's 2018 so I guess just 8 years ago, but it feels like longer, from the before-times), that demonstrated learning brush strokes. At the time there was already a lot of work on GANs, but these are pixel-based methods, and I was really interested in the idea of how to derive descriptive methods of scene generation/understanding. I found this work really interesting because it combined RL and GAN techniques in a creative way. I miss that kind of research.
Now of course VLMs have shown that you can mix modalities in generalized sequence-to-sequence problems and it doesn't surprise me that this kind of thing is possible, but it's so nice to see it done well using modern techniques.
[0] https://proceedings.mlr.press/v80/ganin18a.html
And the fact that it's flowers creates a pretty nice mental model for it! One can plant seeds and cultivate the plants that grow from them but ultimately aren't in complete control of the outcome.
Played a lot with p5js some years ago, might pick it up again and try some of your ideas. The reinforcement learning part sounds about above my skill level though. :)
Wild the possibilities
Are they trained roughly like this? Or is it an LLM conditioned on image? Or on diffusion latents from a model trained to emit SVG-compatible imagery?
People have done plenty with SVGs but it's rare to see human-in-the-loop approaches