in my use sonnet 4.5 seems genuinely more aligned than sonnet 4, even in scenarios which aren't eval flavored. purely anecdotal and not remotely systematic, but it really does seem a lot less inclined to try to pass off failure as success ("the tests are failing, but this is intended behavior!" kind of thing)

the track record of people trying to broadly ensure that humanity continues to be in control of the future

What track record?

But do they also generalize out of training distribution more similarly? If so, why?

Neither of them is going to generalize very well out of distribution, and to the extent they do it will be via looking for features that were present in-distribution. The old adage "to imagine 10-dimensional space, first imagine 3-space, then say 10 really hard".

My guess is that basically every learning system which tractably approximates Bayesian updating on noisy high dimensional data is going to end up with roughly Gaussian OOD behavior. There's been some experiments where (non-adversarially-chosen) OOD samples quickly degrade to uniform prior, but I don't think that's been super robustly studied.

The way humans generalize OOD is not that our visual systems are natively equipped to generalize to contexts they have no way of knowing about, that would be a true violation of no-free-lunch theorems, but that through linguistic reflection & deliberate experimentation some of us can sometimes get a handle on the new domain, and then we use language to communicate that handle to others who come up with things we didn't, etc. OOD generalization is a process at the (sub)cultural & whole-nervous-system level, not something that individual chunks of the brain can do well on their own.

This is also confusing/concerning for me. Why would it be necessary or helpful to have such a large dataset to align the shape/texture bias with humans?

Well it might not be, but you need large datasets to motivate studying large models, as their performance on small datasets like imagenet is often only marginally better.

A 20b param ViT trained on 10m images at 224x224x3 is approximately 1 param for every 75 subpixels, and 2000 params for every image. Classification is an easy enough objective that it very likely just overfits, unless you regularize it a ton, at which point it might still have the expected shape bias at great expense. Training a 20b param model is expensive, I don't think anyone has ever spent that much on a mere imagenet classifier, and public datasets >10x the size of imagenet with any kind of labels only started getting collected in 2021.

To motivate this a bit, humans don't see in frames but let's pretend we do. At 60fps for 12h/day for 10 years, that's nearly 9.5 billion frames. Imagenet is 10 million images. Our visual cortex contains somewhere around 5 billion neurons, which is around 50 trillion parameters (at 1 param / synapse & 10k synapses / neuron, which is a number I remember being reasonable for the whole brain but vision might be 1 or 2 OOM special in either direction).

adversarial examples definitely still exist but they'll look less weird to you because of the shape bias.

anyway this is a random visual model, raw perception without any kind of reflective error correction loop, I'm not sure what you expect it to do differently, or what conclusion you're trying to draw from how it does behave? the inductive bias doesn't precisely match human vision, so it has different mistakes, but as you scale both architectures they become more similar. that's exactly what you'd expect for any approximately Bayesian setup.

the shape bias increasing with scale was definitely conjectured long before it was tested. ML scaling is very recent though,and this experiment was quite expensive. Remember when GPT-2 came out and everyone thought that was a big model? This is an image classifier which is over 10x larger than that. They needed a giant image classification dataset which I don't think even existed 5 years ago.

Scale basically solves this too, with some other additions (not part of any released version of MJ yet) really putting a nail in the coffin, but I can't say too much here w/o divulging trade secrets. I can say that I'm surprised to hear that SD3 is still so much worse than Dalle3, Ideogram on that front - I wonder if they just didn't train it long enough?

They put too much emphasis on high frequency features, suggesting a different inductive bias from humans.

This was found to not be true at scale! It doesn't even feel that true w/weaker vision transformers, seems specific to convnets. I bet smaller animal brains have similar problems.

Order matters more at smaller scales - if you're training a small model on a lot of data and you sample in a sufficiently nonrandom manner, you should expect catastrophic forgetting to kick in eventually, especially if you use weight decay.

I think I can just tell a lot of stuff wrt human values! How do you think children infer them? I think in order for human values to not be viable to point to extensionally (ie by looking at a bunch of examples) you have to make the case that they're much more built-in to the human brain than seems appropriate for a species that can produce both Jains and (Genghis Khan era) Mongols.

I'd also note that "incentivize" is probably giving a lot of the game away here - my guess is you can just pull them out much more directly by gathering a large dataset of human preferences and predicting judgements.

Why do you expect it to be hard to specify given a model that knows the information you're looking for? In general the core lesson of unsupervised learning is that often the best way to get pointers to something you have a limited specification for is to learn some other task that necessarily includes it then specialize to that subtask. Why should values be any different? Broadly, why should values be harder to get good pointers to than much more complicated real-world tasks?

yeah I basically think you need to construct the semantic space for this to work, and haven't seen much work on that front from language modeling researchers.

drives me kinda nuts because I don't think it would actually be that hard to do, and the benefits might be pretty substantial.