Hacker News Reader

Comment by jjcm | Hacker News Reader

Two things stand out to me with this:

1. Drops compute required for training by ~20%. This approach wont just help the ever escalating model sizes larger companies are pushing for, it means things like autoresearch can iterate on new model architectures faster.

2. WAY lower bandwidth requirements for inference. Means with approaches like this it should run on consumer hardware far better. It apparently requires 1/6th the memory bandwidth of a traditional approach for better results.

This is a big improvement if it can be generalized. They're claiming it's a drop in replacement, so it seems like it can as well.

View on HN

> Drops compute required for training by ~20%.

This is not true. Authors claim that w.r.t. training, their method adds negigible overhead for AttnRes with no memory impact (but is way more complicated for Block AttnRes since we need to use pipelining for larger models, hence the O(Ld) & O(Nd) figures, with N ≪ L).

> WAY lower bandwidth requirements for inference.

Also not true. Paper has nothing to do with inference, apart from the benchmarks. If you're looking at the graph about "compute advantage," it's about training compute. They do some interpolation to get to the 1.25x number, basically answering the question "if non-AttnRes architecture were trained, how much compute would it take to get to the same loss as AttnRes?" (The answer being ~20% more compute.) It's an interesting claim, but there's all kinds of weird and unexpected convergence that can happen, so take it with a grain of salt.

I think what they're getting at is that for a given unit of compute, this method achieves 125% performance.

If model A reaches performance level 100 using 100 units of compute using old methods, and you train model B using AttnRes, aiming at performance level 100, it costs you 80 units of compute.

It probably doesn't map precisely, but that's where people are diverging from the claim - it doesn't explicitly say anything about reduced inference or training time, but that's the implicit value of these sorts of things. Less compute to equivalent performance can be a huge win for platforms at scale as well as for local models.

> I think what they're getting at is that for a given unit of compute, this method achieves 125% performance.

This is not what they're getting at; I explained exactly what they're getting at. I mean, your equivalence of "loss" (what authors actually measured) and "performance" is just bizarre. We use benchmarks to measure performance, and the numbers there were like 1-5% better (apart from the GPQA-Diamond outlier).

Do people even read these papers?

> Do people even read these papers?

Overwhelmingly, no. You may have mistaken this for a lab's reading group, but most people here just skim the README, maybe read the abstract or figures. Expecting them to do more is uh... a bit strange?

But also you can forgive people for equating loss with performance, which are admittedly different but related ideas.

> 2. WAY lower bandwidth requirements for inference. Means with approaches like this it should run on consumer hardware far better. It apparently requires 1/6th the memory bandwidth of a traditional approach for better results.

That should be the headline right there. Giant side 60 font headline.

Some people have PhDs in burying the lede!

except it's not true

It's not not true, it's just that things are getting lost in the excitement. There are some specific cases where there's a big boost, it's just not exactly what people are hoping.

>>>The "1/6th" specifically appears in community comparisons to DeepSeek's mHC (multi-lane highway connections, a prior technique for better depth-wise information flow in deep models). Several Chinese-language sources and downstream discussions (e.g., translated articles, YouTube breakdowns, and blogs like houdao.com) state that Block AttnRes achieves comparable (or better) performance to mHC while using only one-sixth of the data read/write volume (or memory bandwidth pressure) during inference/engineering deployment.

There are specific cases where that speedup does occur; it's not going to translate exactly into local models or other architectures or hardware.

No. It seems to me that the comment is objectively incorrect. The original comment was talking about inference and from what I can tell, it is strictly going to run slower than the model trained to the same loss without this approach (it has "minimal overhead"). The main point is that you wont need to train that model for as long.

Reply on news.ycombinator.com