frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Multi-Stream LLMs: new paper on parallelizing/separating prompts, thinking, I/O

https://arxiv.org/abs/2605.12460
22•atomicthumbs•1h ago

Comments

atomicthumbs•1h ago
New paper out of the Max Planck Institute for Intelligent Systems. If this holds up, it seems big.

Abstract: The continued improvements in language model capability have unlocked their widespread use as drivers of autonomous agents, for example in coding or computer use applications. However, the core of these systems has not changed much since early instruction-tuned models like ChatGPT. Even advanced AI agents function on message exchange formats, successively exchanging messages with users, systems, with itself (i.e. chain-of-thought) and tools in a single stream of computation. This bottleneck to a single stream in chat models leads to a number of limitations: the agent cannot act (generate output) while reading, and in reverse, cannot react to new information while writing. Similarly, the agent cannot act while thinking and cannot think while reading or acting on information. In this work, we show that models can be unblocked by switching from instruction-tuning for sequential message formats to instruction-tuning for multiple, parallel streams of computation, splitting each role into a separate stream. Every forward pass of the language model then simultaneously reads from multiple input streams and generates tokens in multiple output streams, all of which causally depend on earlier timesteps. We argue that this data-driven change remedies a number of usability limitations as outlined above, improves model efficiency through parallelization, improves model security through better separation of concerns and can further improve model monitorability.

BBEdit 16

https://www.barebones.com/products/bbedit/bbedit16.html
204•qaz_plm•2h ago•57 comments

Project Hail Mary – Stellar Navigation Chart

https://valhovey.github.io/gaia-mary/
310•speleo•4h ago•87 comments

Flipper One – we need your help

https://blog.flipper.net/flipper-one-we-need-your-help/
938•sandebert•10h ago•392 comments

Blog ran on Ubuntu 16.04 for 10 years. I migrated it to FreeBSD

https://crocidb.com/post/this-blog-ran-on-ubuntu-16-04-for-10-years-i-migrated-it-to-freebsd/
57•speckx•2h ago•11 comments

Show HN: Freenet, a peer-to-peer platform for decentralized apps

https://freenet.org/
98•sanity•6h ago•33 comments

Using Kagi Search with Low Vision

https://veroniiiica.com/using-kagi-search-with-low-vision/
30•speckx•1h ago•1 comments

We're testing new ad formats in Search and expanding our Direct Offers pilot

https://blog.google/products/ads-commerce/google-marketing-live-search-ads/
525•sofumel•11h ago•439 comments

Seattle Shield, an intelligence-sharing network operated by the Seattle police

https://prismreports.org/2026/05/20/seattle-shield-private-companies-surveillance/
355•root-parent•3h ago•146 comments

Was my $48K GPU server worth it?

https://rosmine.ai/2026/05/13/was-my-48k-gpu-worth-it/
154•apwheele•3d ago•117 comments

Indexing a year of video locally on a 2021 MacBook with Gemma4-31B (50GB swap)

https://blog.simbastack.com/indexed-a-year-of-video-locally/
225•asenna•7h ago•77 comments

Spotify will start reserving concert tickets for fans

https://www.hollywoodreporter.com/music/music-industry-news/spotify-will-start-reserving-concert-...
58•elffjs•4h ago•113 comments

Waymo pauses Atlanta service as its robotaxis keep driving into floods

https://techcrunch.com/2026/05/21/waymo-pauses-atlanta-service-as-its-robotaxis-keep-driving-into...
182•mattas•4h ago•230 comments

Python 3.15: features that didn't make the headlines

https://blog.changs.co.uk/python-315-features-that-didnt-make-the-headlines.html
293•rbanffy•10h ago•132 comments

Thoughts on People and Blogs

https://afranca.com.br/thoughts-on-people-and-blogs/
13•meysamazad•2d ago•1 comments

ParadeDB (YC S23) Is Hiring Distributed Systems/Platform Engineers

1•philippemnoel•4h ago

News outlets are limiting the Internet Archive’s access to their journalism

https://www.niemanlab.org/2026/05/more-than-340-local-news-outlets-are-limiting-the-internet-arch...
157•jaredwiener•4h ago•51 comments

Lost Images from the 1945 Trinity Nuclear Test Restored

https://spectrum.ieee.org/trinity-nuclear-test
248•pseudolus•10h ago•78 comments

Where are all the UK red telephone kiosks?

https://www.thek6project.co.uk/
49•Kaibeezy•3h ago•32 comments

Mounting git commits as folders with NFS (2023)

https://jvns.ca/blog/2023/12/04/mounting-git-commits-as-folders-with-nfs/
75•pvtmert•2d ago•40 comments

Google's Antigravity bait and switch

https://www.0xsid.com/blog/antigravity-bait-n-switch
439•ssiddharth•7h ago•226 comments

Launch HN: Runtime (YC P26) – Sandboxed coding agents for everyone on a team

https://www.runtm.com/
44•gustrigos•5h ago•17 comments

Museum of Pocket Calculating Devices

https://www.calculators.de/
52•ohjeez•5h ago•12 comments

Throwing AI-generated walls of text into conversations

https://noslopgrenade.com/
422•napolux•11h ago•253 comments

Chewing gum restores dad's taste and smell years after Covid

https://discover.swns.com/2026/05/chewing-gum-restores-dads-taste-and-smell-years-after-covid/
109•speckx•4h ago•45 comments

Show HN: Agent.email – sign up via curl, claim with a human OTP

28•adisingh13•4h ago•28 comments

Bournegol??? (2014)

https://oldhome.schmorp.de/marc/bournegol.html
26•greyface-•2d ago•4 comments

Vivaldi 8.0

https://vivaldi.com/blog/vivaldi-on-desktop-8-0/
305•OuterVale•13h ago•216 comments

Multi-Stream LLMs: new paper on parallelizing/separating prompts, thinking, I/O

https://arxiv.org/abs/2605.12460
22•atomicthumbs•1h ago•1 comments

Who wins and who loses in prediction markets? Evidence from Polymarket

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6443103
111•vcf•8h ago•101 comments

FatGid: FreeBSD 14.x kernel local privilege escalation

https://fatgid.io/
80•WhyNotHugo•9h ago•36 comments