ICE claim that a man shattered his skull running into wall triggers tension at a Minnesota hospital

Nurses in Minneapolis doubted federal immigration officers' claims about a Mexican immigrant's severe injuries.

apnews.com · AP News

AI가 만든 가짜 논문 인용, 세계 최고 AI 학회도 속았다

지디넷코리아 | 네이버 · 세계 최고 권위의 인공지능(AI) 학회에서 AI가 만들어낸 가짜 논문 인용이 대거 발견됐다. AI 전문가들조차 AI가 생성한 허위 정보를 걸러내지 못하면서 연구 윤리와 학문의 신뢰성에 대한 우려가 커지고 있다. 포천

n.news.naver.com

In Praise of –dry-run

0 comments

lobste.rs · Lobsters

www.usenix.org

Fedify: Building ActivityPub servers without the pain - Slidev

"Implementing ActivityPub looks simple at first—it's just JSON over HTTP,\nright? Then you hit JSON-LD context resolution. Then HTTP Signature\nverification fails on Mastodon but works on Misskey. Then you realize\nthe spec spans hundreds of pages across W3C documents and [FEPs] (Fediverse\nEnhancement Proposals), and every implementation interprets them differently.\n\nI went through this pain building [Hollo], a single-user microblogging server.\nHalfway through, I realized I was building a framework instead of an app.\nSo I extracted that framework and called it Fedify.\n\n[Fedify] is an opinionated ActivityPub framework for TypeScript. It handles\nthe protocol plumbing so you can focus on your application logic.\n\nIn this talk, I'll cover:\n\n - *Type-safe vocabulary:* The Activity Vocabulary spec is loosely defined,\n but Fedify maps it to strict TypeScript types. Your IDE knows that\n `Note.content` is a `LanguageString`, and calling\n `await create.getActor()` returns an `Actor` object.\n No more guessing at property shapes.\n\n - *Comprehensive signature support:* Fedify implements four authentication\n mechanisms—HTTP Signatures (draft-cavage), HTTP Message Signatures (RFC\n 9421), Linked Data Signatures, and Object Integrity Proofs (FEP-8b32).\n For HTTP Signatures, it uses [double-knocking]: trying RFC 9421 first,\n falling back to draft-cavage if rejected, and remembering the preference.\n This kind of interoperability work is exactly what you shouldn't have to\n do yourself.\n\n - *Framework-agnostic design:* Fedify works as middleware for Hono, Express,\n Fastify, Next.js, or any framework that speaks `Request`/`Response`.\n Bring your own database, ORM, and auth—Fedify only needs a key–value store\n for caching.\n\n - *CLI toolchain:* The `fedify inbox` command spins up an ephemeral server\n to receive and inspect activities. `fedify lookup` fetches any ActivityPub\n object by URL or fediverse handle—including from servers that require\n [authorized fetch]. No need to create throwaway accounts on production\n instances.\n\nI'll also share production stories: [Ghost chose Fedify] for federating their\npublishing platform rather than implementing the protocol themselves. Hollo\ndemonstrates single-user microblogging with full Mastodon API compatibility.\n[Hackers' Pub] shows how a developer community can integrate with\nthe fediverse.\n\nWhether you're building a new federated service or adding ActivityPub to\nan existing app, this talk will show you how Fedify turns months of protocol\nwrangling into days of actual development.\n\n[FEPs]: https://w3id.org/fep/\n[Hollo]: https://docs.hollo.social/\n[Fedify]: https://fedify.dev/\n[double-knocking]: https://swicg.github.io/activitypub-http-signature/#how-to-upgrade-supported-versions\n[authorized fetch]: https://swicg.github.io/activitypub-http-signature/#authorized-fetch\n[Ghost chose Fedify]: https://activitypub.ghost.org/day-4/\n[Hackers' Pub]: https://hackers.pub/\n"

hongminhee.codeberg.page

omu (@boriborisennbei)

チームみらいの賛否が色々出ている。 比例で選ぶ予定の人は、投票前にこれを見て、この高山さとし幹事長に税金を使って欲しいなと思えたら投票すればいいですよ。 コンサルを経てAIスタートアップ3社の事業開発（いずれも企業名は不明）に携わってきた38歳。この仕上がり。 #比例はみらい #衆院選

nitter.net · Nitter

(3ページ目)24歳の元AKBメンバーが話題に…「日韓のファンがザワついた」SAY MY NAMEを日本人が注目すべき“3つの理由” | 文春オンライン

(3ページ目)K-POPになじみがない人もハマる音楽性 SAY MY NAMEの楽曲はこれまでK-POPを聴いてこなかった人、あまり馴染みがない人にも自信を持ってオススメしたい。音楽ジャンルは多岐に渡るが、全体的に…

bunshun.jp · 文春オンライン

The best Mastodon client now has an iOS version! - Matt Fantinel

Meet iPhanpy, the best way to use Phanpy on iOS.

fantinel.dev

Demystifying ARM SME to Optimize General Matrix Multiplications

General Matrix Multiplication (GEMM) is a critical kernel in high-performance computing and deep learning. While modern architectures like ARM's Scalable Matrix Extension (SME) introduce dedicated hardware for matrix operations, existing linear algebra libraries fail to fully exploit its potential, particularly for large matrices. This paper presents MpGEMM, an open-source library that leverages key architectural features of SME to optimize GEMM across multiple precisions. Through a systematic characterization of SME, we derive optimization guidelines that inform our design. MpGEMM employs cache-aware partitioning, efficient data packing with on-the-fly transposition, and specialized micro-kernels that utilize multi-vector loads and all available tile registers. Evaluated on an Apple M4 Pro with real-world workloads from DeepSeek and LLaMA, MpGEMM achieves an average speedup of 1.23x over the vendor-optimized Apple Accelerate library and significantly outperforms other open-source alternatives.

arxiv.org · arXiv.org

fep/fep/c390/fep-c390.md at main

fep - Fediverse Enhancement Proposals

codeberg.org · Codeberg.org