Given the size of your team, have y'all considered going into niche like making a DLSS 5 competitor

Slightly off-topic but thoughts on this https://arxiv.org/pdf/2410.08159

Just skimmed the paper (haven't seen this before). Might need to read more carefully, but at first glance I don't understand their intuition why the "Markovian property limits the model’s ability to fully utilize the generation trajectory".

When it comes to multi-resolution training (e.g. matryoshka training), there are precedents that don't require this AR formulation.

They cite MAR (https://arxiv.org/pdf/2406.11838), which I think is a much clearer articulation of "autoregressive diffusion". MAR uses an autoregressive base (like an LLM) and staples on a small MLP on-top that's trained as a diffusion head. That makes more sense to me, since you can leverage the "knowledge prior" from an LLM and have it generate images. That's probably how nano-bannana and GPT-Image broadly work.

Zooming out, it's not clear to me from any of the work WHY autoregressive diffusion on it's own is better than regular diffusion. Most autoregressive diffusion models use speculative decoding, because pure autoregressive diffusion is too slow to run at inference time.

The one ~magical~ thing about auto-regressive diffusion (IMO) has nothing to do with autoregressive vs. fully-bidirectional diffusion. But simply, the face that you can do it on-top of a LLM. That means the LLM can look at it's generation, assess it's quality, think on what's broken, and then call itself to edit the image and fix it. With methods like RLVR, that means you can essentially guarantee the correctness of your image (along the axes you've RLVR-ed).

In my opinion its a great idea because you get reuse hidden states which allows you to do something akin to thinking/[online learning]. If you only evolve use input space outputs you're dealing with more decoding pressure.

Also related to what I was saying see these:

https://arxiv.org/pdf/2609.16372v1

https://arxiv.org/pdf/2609.01449v1 (this one is quite mindblowing cause they noise they the actual input space every time but it still works)

https://arxiv.org/pdf/2609.11801

Thanks for the references! I’ll have to take a deeper look, haven’t read these before.

There are better ways to apply SVD to images than the way its usually taught

https://www.youtube.com/watch?v=ZGwVlnuuzt4

The HN poster of this owns an extremely anti-social page btw https://boomerdeathwatch.com/ . Before he claims that "they deserved it", note how the page itself does nothing but make them look like a weirdo lol

I've become more and more convinced that stuff like this is from state actors trying to divide the West or at least cause some kind of chaos.

I actually strongly disagree, I think this is the natural tendency of many losers and there is a disproportionate amount of them on the internet

both can be true and probably is

Yeah a lot of people have failed to internalize a lot of the meta-lessons of current culture wars

Here's another meta-lesson: "woke" turns out to be broadly popular and is not actually going away.

Yes, many people *genuinely* give a fuck that tech leaders are promoting white supremacy and ethnic cleansing, and there will be actual consequences for this.

> there will be actual consequences for this.

Not implausible but what exactly because sometimes when people throw that phrase its just aggrieved cope

"Woke" is the 20 side of myriad 20-80 issues. It cannot accurately be described as "broadly popular."

I really want to like ai music but none of them seem to be trained on reward models that reward "ambiance" or any sort of interesting sound design (which this paper obviously doesn't even concern), which might be reflective of the people training them not having niche music tastes

Why do you want to like it?

Because there is a shortage of songs I like. I loop the same songs for weeks until I can find a new song

I'm the opposite of you. I don't do songs, I do albums.

And there are just too many albums to go through each week. I'm on Redacted and Orpheus though, pulling the Top 10s each week.

Why not read some music blogs or go to a record store, or talk to your friends about the music they like?

I don't share my friend's music taste and record stores for sure don't have what I like (its usually small soundcloud accounts). Also the things I like about a song are not genre bound, its usually very subtle things that are unsearchable, hence even of the songs I like I usually only like ~30% of the song itself. With a personally tuned reward model I can strictly focus on amplifying the subtleties instead of searching by means of exhausting indirection

The figure on page 5 in [1] is pretty insane

[1] https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/...

The scifi pov has a good track record as this point, you people gotta be more open minded

whoops lol

Smart fridges are a pathological virus, yes /jk

Even normal fridges are a memetic virus, fridges clearly live in a symbiotic relationship with humans. They make our lives better and in return we create more fridges. If they weren't useful to us, we'd stop making more fridges. Any medieval human who sees a fridge would instantly recognize why it's useful and valuable.

OP is right, anything popular is virus-like. Richard Dawkins invented the word meme for this all the way back in 1976.