お疲れ様でした

よし！！！！、！！！

おおおおおん

ワールドカップ見るために起きた

Wasmコンポーネントモデル 1.0に向けた話が出ていてかなりあつい bytecodealliance.org/article…

IEEE COMPSACでスペイン行くで〜〜 Cross-OS Container Transplantation: A Lightweight Defense Against Kernel and Network Exploits – Kotaro Sakamoto, Yuki Nakata, Katsuya Matsubara, Soma Sakaguchi and Shintaro Suzuki ieeecompsac.computer.org/202…

締切のないジャーナル、一生書き直して終わらない病にかかる

KubeCon Japan行きます

月末に妻とシアトル・バンクーバー行く予定だけど、どこ行こうかってなってる（飛行機とホテル抑えて満足してる）

MLSys 2026にて今年新設されたIndustry Trackで発表したレポートをブログにまとめました。 research.sakura.ad.jp/blog/m…

オーラル発表に用いたスライド資料はこちら。 speakerdeck.com/yuukit/sakur…

国際会議MLSys2026で登壇してきました！久しぶりの英語での登壇に加えて、会場に数百人はいたのでさすがに緊張しました。 research.sakura.ad.jp/blog/m…

日本語、台湾華語、英語の辞書を買った。2カ国語を同時に勉強している身としては大変便利

台湾に行きたいわん...🫠

Introducing DiffusionBlocks: Block-wise Neural Network Training via Diffusion Interpretation pub.sakana.ai/diffusionblock… 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: arxiv.org/abs/2506.14202 GitHub: github.com/SakanaAI/Diffusio… 🐟

@opentelemetry is officially a CNCF graduated project! 🎓🎉 OpenTelemetry has become the trusted de facto observability standard, backed by 12,000 contributors from 2,800 organizations and helping teams gain better visibility across distributed systems. Congrats to this incredible community! Read more about the milestone here: bit.ly/4fvcHAb

Today at #MLSys2026, I will be presenting our work on "SAKURAONE: An Open Ethernet-Based AI HPC System and Its Observed Workload Dynamics in a Single-Tenant LLM Development Environment." Join us at Poster Session 3 (Thu 21 May, 6:00pm~8:00pm). Paper: arxiv.org/abs/2604.13600

博士課程の中間発表で🥺ってなってきた

函館に居ます

脳内で歌がループしてる 官方最新版-乖乖歌健康操MV youtu.be/iduQOn91a4o?si=f4xG…