@raiders@darktundra.xyz2026-01-18 15:47:47
How Raiders Are Set to Get Ultimate Evaluation of Fernando Mendoza https://www.si.com/nfl/raiders/onsi/las-vegas-set-ultimate-evaluation-fernando-mendoza
@jby@ecoevo.social2025-12-10 14:17:06
A range-spanning resurrection study of scarlet monkeyflower populations finds relatively little adaptive response to a seven-year drought event, showing that evolutionary rescue is not as predictable as we might hope
https://doi.org/10.1086/738434
@cosmos4u@scicomm.xyz2025-12-01 21:06:04
Assembling GW231123 in Star Clusters through the Combination of Stellar Binary Evolution and Hierarchical Mergers: #BlackHole mergers: https://www.gssi.it/communication/news-events/item/26044-gssi-astrophysicists-reveal-the-role-of-star-clusters-in-black-hole-mergers
@aredridel@kolektiva.social2025-11-28 16:19:02
Ugh why is this always the way. I evaluated like 25 authentication servers for a small scale web project — I do want to support things like OIDC and Passkeys, so this is not something I really want to make myself like the old days of “use crypt() on the passwords and just make a simple database”.
5 of them are just dev mode garbage that will never see the light of day as a thing people use.
2 of them are home network nonsense for people who want enterprise login for their family, but where One Nerd controls the whole user-list.
15 of them are freemium "open source" where they withhold features for their enterprise tier and make them so unfortunately difficult to deploy, all requiring postgresql databases and a complex containerization setup and helm charts and oh so much.
and then there's kanidm, which is great except its opinions make it completely unusable for a community project, it's really more trying to fit the ‘enterprise unix authentication' space. Kudos to them for communicating it but it's the wrong tool, even if it is really good.
And then there's rauthy. Which is exactly what I want, well built and delightful, uses a lightweight embedded database, and even has a peer-to-peer sync for scalability. But customizing it is going to be a lesson in building it from source repeatedly, and its configuration is just a bit strange, and its frontend is extremely Backend Developer Wrote A Web UI. I guess I got a second project. And maybe a third to make debian packages of it.
Yet it really is the best of the options _by far_.
NLNet supported projects continue to punch above their weight class.
@michabbb@social.vivaldi.net2025-12-25 22:05:58
• Type Coercion: 0.041ms (32,600 ops/second)
• Basic Parsing: 0.133ms (16,900 ops/second)
• Large Payloads (10KB): 2.3ms (437 ops/second)
💰 Cost: Free plugin with negligible latency impact (<1ms for typical responses)
🎯 Scope & limitations:
• Fixes JSON syntax errors, not schema adherence
• Works for non-streaming requests only
• XML healing available on request
• Tool calling has few structural issues, schema adherence evaluation coming soon<…
@arXiv_csLG_bot@mastoxiv.page2025-12-22 10:33:20
Can You Hear Me Now? A Benchmark for Long-Range Graph Propagation
Luca Miglior, Matteo Tolloso, Alessio Gravina, Davide Bacciu
https://arxiv.org/abs/2512.17762 https://arxiv.org/pdf/2512.17762 https://arxiv.org/html/2512.17762
arXiv:2512.17762v1 Announce Type: new
Abstract: Effectively capturing long-range interactions remains a fundamental yet unresolved challenge in graph neural network (GNN) research, critical for applications across diverse fields of science. To systematically address this, we introduce ECHO (Evaluating Communication over long HOps), a novel benchmark specifically designed to rigorously assess the capabilities of GNNs in handling very long-range graph propagation. ECHO includes three synthetic graph tasks, namely single-source shortest paths, node eccentricity, and graph diameter, each constructed over diverse and structurally challenging topologies intentionally designed to introduce significant information bottlenecks. ECHO also includes two real-world datasets, ECHO-Charge and ECHO-Energy, which define chemically grounded benchmarks for predicting atomic partial charges and molecular total energies, respectively, with reference computations obtained at the density functional theory (DFT) level. Both tasks inherently depend on capturing complex long-range molecular interactions. Our extensive benchmarking of popular GNN architectures reveals clear performance gaps, emphasizing the difficulty of true long-range propagation and highlighting design choices capable of overcoming inherent limitations. ECHO thereby sets a new standard for evaluating long-range information propagation, also providing a compelling example for its need in AI for science.
toXiv_bot_toot
@nfdi4culture@nfdi.social2025-11-03 10:43:10
Wie gut ist eigentlich eure #GLAM -Sammlung #galerie, #bibliothek, #archiv,
