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!
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.
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. :)
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.
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.
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.
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
Wild the possibilities
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. :)
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.
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