切片即足够:迈向通用的单侧分布式MatMul
Slicing Is All You Need: Towards a Universal One-Sided Distributed MatMul

原始链接: https://arxiv.org/abs/2510.08874

arXivLabs是一个框架,允许合作者直接在我们的网站上开发和分享新的arXiv功能。个人和与arXivLabs合作的组织都认同并接受我们开放、社群、卓越和用户数据隐私的价值观。arXiv致力于这些价值观,并且只与秉持这些价值观的合作伙伴合作。您是否有为arXiv社群增加价值的项目想法?了解更多关于arXivLabs的信息。

A recent arXiv paper titled “Slicing Is All You Need” sparked a humorous discussion on Hacker News, primarily revolving around the overuse of the phrase “is all you need” in academic titles. Many commenters pointed to a previous HN discussion labeling the phrase “harmful,” with some playfully suggesting alternatives like “goto is all you need” or referencing Beatles lyrics. Others advocated for arXiv to actively block submissions using the phrase, alongside similarly versioned model names like “YOLO-v15.” The conversation also touched on the broader trend of researchers emulating popular AI paper titles for perceived prestige, and the hope that these title fads will become amusing historical footnotes in internet culture. Ultimately, the thread highlights a lighthearted frustration with repetitive naming conventions in the AI research community.
相关文章

原文

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

联系我们 contact @ memedata.com