It’s really hard getting dark skin sometimes. A lot of the time it’s not even just the model, LoRAs and Textual Inversions make the skin lighter again so you have to try even harder. It’s going to take conscious effort from people to tune models that are inclusive. With the way media is biased right now, I feel like it’s going to take a lot of effort.
Comment on Why is AI Pornifying Asian Women?
Gaywallet@beehaw.org 11 months agoYou’re absolutely correct, yet ask someone who’s very pro AI and they might dismiss such claims as “needing better prompts”. Also many people may not be as tech informed as you are, and bringing light to algorithmic bias can help them understand and navigate the world we now live in.
Even_Adder@lemmy.dbzer0.com 11 months ago
jarfil@beehaw.org 11 months ago
“Inclusive models” would need to be larger.
Right now people seem to prefer smaller quantized models, with whatever set of even smaller LoRAs on top, that make them output what they want… and only include more generic elements in the base model.
Muehe@lemmy.ml 11 months ago
“Inclusive models” would need to be larger.
[citation needed]
To my understanding the problem is that the models reproduce biases in the training material, not model size. Alignment is currently a manual process after the initial unsupervised learning phase, often done by click-workers (Reinforcement Learning from Human Feedback, RLHF), and aimed at coaxing the model towards more “politically correct” outputs; But ultimately at that time the damage is already done since the bias is encoded in the model weights and will resurface in the outputs just randomly or if you “jailbreak” enough.
In the context of the OP, if your training material has a high volume of sexualised depictions of Asian women the model will reproduce that in its outputs. Which is also the argument the article makes. So what you need for more inclusive models is essentially a de-biased training set designed with that specific purpose in mind.
I’m glad to be corrected here, especially if you have any sources to look at.
Even_Adder@lemmy.dbzer0.com 11 months ago
I wouldn’t mind. I’m here for it.
jarfil@beehaw.org 11 months ago
Are you ready to run a 100B FP64 parameter model? Or even a 10B FP32 one?
Over time, I wouldn’t be surprised if 500B INT8 models became commonplace with neuromorphic RAM, but there’s still some time for that to happen.
helenslunch@feddit.nl 11 months ago
If the author doesn’t know the answer, then it is helpful to provide it. If they know the answer, then why are they phrasing the title as a question?
MBM@lemmings.world 11 months ago
If you genuinely don’t know: because it’s an attention-grabbing title (which isn’t inherently bad)