Krispychip

[Krispychip @krispywe]

Now, ComicTL has finally released the stable version. You can now choose to fully run it locally or use Gemini in cloud mode. I also made a website for it, you can read it at https://t.co/CRAXTFijfx.

[Sakana AI @SakanaAILabs]

Introducing DiffusionBlocks: Block-wise Neural Network Training via Diffusion Interpretation

https://t.co/c9AvsRKybj

What if we didn’t have to hold an entire neural network in memory to train it?

Standard neural net training optimizes all parameters jointly. As a result, the memory required during training grows linearly with the depth of the network.

In our #ICLR2026 paper, we propose DiffusionBlocks, a principled framework to train networks one block at a time, drastically reducing memory requirements while matching end-to-end performance.

With DiffusionBlocks, we split the network into blocks and train them one at a time, so you only need memory for a single block.

How? We explicitly assign each block a role: to move the representation a little closer to the target than the block before it did. That role turns out to be precisely what a diffusion model does, step by step. Each block only needs to optimize its own objective and can be trained independently.

We validated this across five different architectures:

• ViT
• DiT
• Masked diffusion
• Autoregressive transformers
• Recurrent-depth transformers

In each case, performance is competitive with end-to-end training while using a fraction of the memory.

This perspective also extends naturally to recurrent-depth (Looped) transformers, which apply the same network iteratively and normally require expensive backpropagation through time (BPTT). Viewed through DiffusionBlocks, we can replace those multiple iterations with a single forward pass during training.

Read our paper and code, to learn more.

Paper: https://t.co/CRj96VGYQn
GitHub: https://t.co/eNW0K9Xh8E
🐟

[Krispychip @krispywe]

@gavinrbrown1 Lmao that's good one

[Krispychip @krispywe]

@iyoushetwt Hope you're not wasting your time doin Linux ricing

[Krispychip @krispywe]

@iRainbowsaur @Birchlabs I got u bro. Almost every time I read papers, they are always invent new word for something already exist

[Krispychip @krispywe]

@twtayaan No wonder linux user usually not in a relationship, especially Arch guys

[Krispychip @krispywe]

Ah, CVE-2026-31431. For the first time, I might be rethinking Linux and AI. Damn, it 732 Bytes of Python can make you access root access from user. Fk insane founding.

[Krispychip @krispywe]

Not only Mangadex, but you could also use it in X
https://t.co/g1RVt66Gb5

[fanks @fanks_carol]

@Animex_Tweet 加工された翻訳だ
これが本物です

S「このあたりになにか用事でも？」
A「ただの通りすがりだ。君こそなぜここに」
S「こちらの児童館で、時折手伝いを、今日は催し物があるのです」

[Krispychip @krispywe]

Now, ComicTL has finally released the stable version. You can now choose to fully run it locally or use Gemini in cloud mode. I also made a website for it, you can read it at https://t.co/CRAXTFijfx.

[Krispychip @krispywe]

Finally, the long project that I almost forgot got into beta release. You can try it by downloading the artifact from GitHub. https://t.co/w9IefZCXYe. Currently, it is only available with Gemini translation. The stable release will try to include the local LLM as a translator.

[Krispychip @krispywe]

@pnkrtn_ @rayzhudev @catboosted wait a month is serious gap?

[Krispychip @krispywe]

@necrosamus I hear that a bit long time ago, and the @idwiki already post that

[Krispychip @krispywe]

Bruh I just upload PDF and it just vommit its own think and endless loop like I do prompt attack @GeminiApp

[Krispychip @krispywe]

@cneuralnetwork They're too innocent for this cruel world

[Krispychip @krispywe]

@julioleiva2121 @BrunsJulian1541 @ChatgptLunatics Bro got trigger immediately just because he talk about model type

[Krispychip @krispywe]

LinkedIn is a positively toxic echo chamber. Everyone licks each other's boots in the comments, saying "nice insight" because they are too afraid to correct wrong math. Politeness is just a mask for incompetence. Stop trusting vibes and start reading the docs. (12/12)

[Krispychip @krispywe]

This is why I hate LinkedIn. Too many people with glorified titles but they confidently post trash because they can’t even read properly. An "AI Engineer" promoting a custom training loop for imbalance that is actually tutorial how to sabotage your own model. (1/12)

[Krispychip @krispywe]

But because Focal Loss doesn't look as "complex" as a 40-line custom filtering pipeline, people think it is less professional. Complexity is not a feature; it is a bug. A senior engineer deletes code; they don't add unreadable dataset mutilation. (11/12)