Why is this seemingly official Google blog post on this random non-Google domain?
Not just that ... Google Sheets must be the all-time worst way to distribute 200 short strings.
Imagen, a text-to-image diffusion model
201–210 of 661 posts
Re: Imagen, a text-to-image diffusion model
#202Earlier quoted context omitted.
While appspot.com is a Google domain, anyone can register domains under it. It would be similarly surprising to see an official GitHub blog post under someproject.github.io
You mean like: https://say-can.github.io/ This is common in the research PA. People don't want to deal with broccoli man [1]. [1] https://www.youtube.com/watch?v=3t6L-FlfeaI
Looking at that link, I don't think that is a GitHub publication? It is marked Robotics at Google and Everyday Robotics.
Re: Imagen, a text-to-image diffusion model
#203Earlier quoted context omitted.
At the end of a day, if you ask for a nurse, should the model output a male or female by default? If the input text lacks context/nuance, then the model must have some bias to infer the user's intent. This holds true for any image it generates; not just the politically sensitive ones. For example, if I ask for a picture of a person, and don't get one with pink hair, is that a shortcoming of the model? I'd say that bi…
> At the end of a day, if you ask for a nurse, should the model output a male or female by default? Randomly pick one. > Trying to generate a model that's "free of correlative relationships" is impossible because the model would never have the infinitely pedantic input text to describe the exact output image. Sure, and you can never make a medical procedure 100% safe. Doesn't mean that you don't try to make them safe…
what if I asked the model to show me a sunday school photograph of baptists in the National Baptist Convention?
Re: Imagen, a text-to-image diffusion model
#204>While we leave an in-depth empirical analysis of social and cultural biases to future work, our small scale internal assessments reveal several limitations that guide our decision not to release our model at this time. Some of the reasoning: >Preliminary assessment also suggests Imagen encodes several social biases and stereotypes, including an overall bias towards generating images of people with lighter skin tones…
In short, the generated images are too gender-challenged-challenged and underrepresent the spectrum of new normalcy!
Re: Imagen, a text-to-image diffusion model
#205>While we leave an in-depth empirical analysis of social and cultural biases to future work, our small scale internal assessments reveal several limitations that guide our decision not to release our model at this time. Some of the reasoning: >Preliminary assessment also suggests Imagen encodes several social biases and stereotypes, including an overall bias towards generating images of people with lighter skin tones…
This seems bullshit to me, considering Google translate and google images encode the same biases and stereotypes, and are widely available.
yea but now they aren't giving people more data-points to attack them with such nonsense arguments.
Re: Imagen, a text-to-image diffusion model
#206>While we leave an in-depth empirical analysis of social and cultural biases to future work, our small scale internal assessments reveal several limitations that guide our decision not to release our model at this time. Some of the reasoning: >Preliminary assessment also suggests Imagen encodes several social biases and stereotypes, including an overall bias towards generating images of people with lighter skin tones…
So glad the company that spies on me and reads my email for profit is protecting me from pictures that don't look like TV commercials.
Gmail doesn’t read your email for ads anymore. They read it to implement spam filters, and good thing too. Having working spam filters is indeed why they make money though.
Re: Imagen, a text-to-image diffusion model
#207>While we leave an in-depth empirical analysis of social and cultural biases to future work, our small scale internal assessments reveal several limitations that guide our decision not to release our model at this time. Some of the reasoning: >Preliminary assessment also suggests Imagen encodes several social biases and stereotypes, including an overall bias towards generating images of people with lighter skin tones…
This raises some really interesting questions. We certainly don't want to perpetuate harmful stereotypes. But is it a flaw that the model encodes the world as it really is, statistically, rather than as we would like it to be? By this I mean that there are more light-skinned people in the west than dark, and there are more women nurses than men, which is reflected in the model's training data. If the model only gener…
It’s the same as with an artist: “hey artist, draw me a nurse.” “Hmm okay, do you want it a guy or girl?” “Don’t ask me, just draw what I’m saying.” The artist can then say: “Okay, but accept my biases.” or “I can’t since your input is ambiguous.”
For a one-shot generative algorithm you must accept the artist’s biases.
Re: Imagen, a text-to-image diffusion model
#208Really impressive. If we are able to generate such detailed images, is there anything similar for text to music? I would I though that it would be simpler to achieve than text to image.
SymphonyNet: https://youtu.be/m4tT5fx_ih8
Re: Imagen, a text-to-image diffusion model
#209Metacalculus, a mass forecasting site, has steadily brought forward the prediction date for a weakly general AI. Jaw-dropping advances like this, only increase my confidence in this prediction. "The future is now, old man." https://www.metaculus.com/questions/3479/date-weakly-general...
I don't see how this gets us (much) closer to general AI. Where is the reasoning?
I think this serves at least as a clear demonstration of how advanced the current state of AI is. I had played with GPT-3 and that was very impressive but I couldn't even dream something as good as D-ALLE 2 was already possible.
Re: Imagen, a text-to-image diffusion model
#210Earlier quoted context omitted.
Additionally, if you optimize for most-likely-as-best, you will end up with the stereotypical result 100% of the time, instead of in proportional frequency to the statistics. Put another way, when we ask for an output optimized for "nursiness", is that not a request for some ur stereotypical nurse?
You could stipulate that it roll a die based on percentage results - if 70% of Americans are "white", then 70% of the time show a white person - 13% of the time the result should be black, etc. That's excessively simplified but wouldn't this drop the stereotype and better reflect reality?
Is this going to be hand-rolled? Do you change the prompt you pass to the network to reflect the desired outcomes?