I think your excitement is justified. The paper is claiming a serious bridge between AR quality and parallel decoding, and the lossless LoRA-assisted mode is the wildest part.

I don't understand how you can compare against the base model output without generating with the base model, in which case what's the point?

Eh. There is nothing diffusion about this. Nothing to do with denoising. This setup is still purely causal, making it quite a dishonest framing IMO. There is no more introspection here than what happens in MTP + SD setups.

Let me explain what is going on here. This is basically a form of multi-token prediction. And speculative decoding in inference. See my earlier post[1] to understand what that is. TL;DR, in multi-token prediction you train separate LM heads to predict the next as well as next to next token as well as... Upto chosen next kth token. Training multiple LM heads is expensive and can be unnecessary, so what people typically do is have a common base for all the k heads, explained further in [1]. These guys do another variant.

Here is what they do mechanically, given a sequence p consisting of five tokens PE([p1, p2, p3, p4, p5]). Where PE(.) adds relative position info to each token.

1. Create an augmented sequence PE([p1 MASK MASK MASK MASK]). Do a training pass on that, with the ground truth sequence p1..5. Here it is trained to, for example, to predict p3 given p1+pos=-2 MASK+pos=-1 MASK+pos=0, loosely notating.

2. Then separately[2], train it as usual on PE([p1 p2 p3 p4 p5]).

Step (1) teaches it to do multi-token prediction, essentially the single LM head will (very very loosely speaking) condition on the position `k` of the special MASK token and "route" it to the "implicit" k'th LM head.

Step (2) teaches it to be a usual LLM and predict the next token. No MASK tokens involved.

So far, you have trained a multi-token predictor.

Now during inference

You use this for speculative decoding. You generate 5 tokens ahead at once with MASK tokens. And then you run that sequence through the LLM again. This has the same benefits as usual speculative decoding, namely that you can do matrix-matrix multiplication as opposed to matrix-vector. The former is more memory-bandwidth efficient due to higher arithmetic intensity.

here is an example,

query = ["what", "is", "2+2"]) prompt = PE([...query, MASK*5]) you run output = LLM(prompt). Say output is ["what", "is", "2+2", "it", "is", "4"]. Note that the NN is trained to predict the kth next token when faced with positionally encoded MASK tokens. So you get all 5 in one go. To be precise, it learns to predict "4" given ["what", "is", "2+2", MASK, MASK]. Since it does not need the "it" and "is" explicitly, you can do it in parallel with generating the "it" and the "is". "is" is predicted given ["what", "is", "2+2", MASK], for example, and that also doesn't depend on the explicit "it" being there, and thus can also be done in parallel with generating "it", which is just normal generating the next token given the query. And then you use this as a draft in your speculative decoding setup.

Their claim is that using a multi-token predictor this way as a draft model works really well. To be clear, this is still causal, the reason diffusion models have hype is because they are capable of global refinement. This is not. In the same thread as [1], I explain how increasing the number of MASK tokens, i.e increasing `k`, i.e the number of tokens you predict at once in your multi-token prediction setup quickly leads to poor quality. This paper agrees with that. They try out k=2,3,4,8. They see a drop in quality at 8 itself. So finally, this is 4-token-prediction with self-speculative decoding(sans LayerSkip or such), removing seemingly no existing limitation of such setups. It is definitely an interesting way to train MTP though.

[1] https://news.ycombinator.com/item?id=45221692

[2] Note that it is computationally a single forward pass. Attention masks help you fuse steps 1 and 2 into a single operation. However, you still have 2 separate loss values.

https://www.inceptionlabs.ai/

This startup seems to have been at it a while.

From our look into it - amazing speed, but challenges remain around time-to-first-token user experience and overall answer quality.

Can absolutely see this working if we can get the speed and accuracy up to that “good enough” position for cheaper models - or non-user facing async work.

One other question I’ve had is wondering if it’s possible to actually set a huge amount of text to diffuse as the output - using a larger body to mechanically force greater levels of reasoning. I’m sure there’s some incredibly interesting research taking place in the big labs on this.

It's being explored right now for speculative decoding in the local-LLM space, which I think is quite interesting as a use-case

https://www.emergentmind.com/topics/dflash-block-diffusion-f...

I have. It requires a distinct intuition compared to a normal language model. Very well suited to certain problems.

I've found the latency and pricing make Mercury 2 extremely compelling for some UX experiments focused around automated note tagging/interlinking. Far more than the Gemini Flash Lite I used before, it made some interactions nearly frictionless, very close to how old school autocomplete/T9/autocorrect works in a manner that users don't even think about the processes behind it.

Sadly, it does not perform at the level of e.g. Haiku 3.5 for tool calling, despite their own benchmarks claiming parity with Haiku 4.5, but it does compete with Flash Lite there too.

Anything with very targeted output, sufficient existing input and that benefits from a seamless feeling lends itself to dLLMs. Could see a place in tab-complete too, though Cursors model seems to be sufficiently low latency already.

I've been playing with a Swift implementation of a diffusion language model (WeDLM), but performance is not yet acceptable and it still generates roughly from left-to-right like a language model (just within a sliding window rather than strictly token-by-token... but that doesn't matter when the sliding window is only like 16 tokens.)

Well, you can take the output of a first pass and pass it back through the model like AR “reasoning” models do at inference time.

Block auto regressive generation can give you big speedups.

Consider that outputting two tokens at a time will be a (2-epsilon)x speedup over running one token at a time. As your block size increases, you quickly get to fast enough that it doesn't matter sooooo much whether you're doing blocks or actual all-at-once generation. What matters, then, is there quality trade-off for moving to block-mode output. And here it sounds like they've minimized that trade-off.

[dead]

It's clearly a typo on the year, April 12 was two days ago, a quick check in HuggingFace shows that they were uploaded 5 days ago.

This translates to

> I understand it improved by 3x, but has the bottleneck shifted from Memory Bandwidth to Compute? Or is Memory Bandwidth still dominant?

But why did you post your comment in Japanese? We have so many good options for automated translation nowadays!

でも、なぜ日本語でコメントを投稿したんですか？最近は自動翻訳の良い選択肢がたくさんあるのに！

I'm not in on the joke, can someone ELI5