We’re releasing highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights. Depending on the chosen sparsity, these kernels can run orders of magnitude faster than cuBLAS or cuSPARSE.
Summary generated by The Flow from the publisher's feed. The full article lives at OpenAI Blog.
arXiv:2607. 08786v1 Announce Type: cross Abstract: With the growing deployment of large language models (LLMs), LLM inference cost has become a key challenge.
arXiv:2606. 02328v1 Announce Type: new Abstract: We explore Riemannian optimization techniques for rank-factored matrix parameters, targeting contemporary deep learning applications.
We’ve developed an unsupervised system which learns an excellent representation of sentiment, despite being trained only to predict the next character in the text of Amazon reviews.
arXiv:2606. 06742v1 Announce Type: new Abstract: TorchKM is an open-source library for kernel machines, including support vector machines, kernel logistic regression, and kernel quantile regression, with GPU acceleration.
arXiv:2605. 26632v3 Announce Type: replace Abstract: Diffusion Transformers (DiT) achieve strong performance in image generation but incur substantial inference costs.
arXiv:2605. 26632v2 Announce Type: replace Abstract: Diffusion Transformers (DiT) achieve strong performance in image generation but incur substantial inference costs.