Tootfinder

Helsinki-based DataCrunch, which aims to become Europe's first AI cloud hyperscaler, raised a €55M Series A, bringing its total funding to €76.5M (Tamara Djurickovic/Tech.eu)
https://tech.eu/2025/09/08/datacrunch-raises-eur5…

DataCrunch raises €55M to boost EU AI sovereignty with green cloud infrastructure
Helsinki-based DataCrunch, a provider of high-performance AI infrastructure, has raised €55 million in a Series A round to scale its compute platform and advance its ambition to become Europe’s first AI cloud hyperscaler.The round was led by byFounders, Skaala, Varma pension fund, and Tesi, with participation from J12 Ventures and angel investors. Debt financing came from Nordea and Armada Credit Partners, alongside Danske Bank, Norion Bank, and Local Tapiola. With this raise, DataCru…

Lin-Lu-Yau Ricci curvature on hypergraphs
Yulu Tian, Liang Zhao
https://arxiv.org/abs/2507.04109 https://arxiv.org/pdf/2507.04109

Lin-Lu-Yau Ricci curvature on hypergraphs
In this paper, we introduce a unified framework for defining Lin-Lu-Yau (LLY) Ricci curvature on both undirected and directed hypergraphs. By establishing upper bounds and monotonicity properties for the parameterized curvature $κ_α$, we justify the well-posedness and compatibility of our definitions. Furthermore, we prove a Bonnet-Myers-type theorem for hypergraphs, which highlights the potential of LLY Ricci curvature in hypergraph analysis, particularly in studying geometric and structural…

Transitivity Preserving Projection in Directed Hypergraphs
Eric Parsonage, Matthew Roughan, Hung X Nguyen
https://arxiv.org/abs/2509.04543 https://arxiv.or…

Transitivity Preserving Projection in Directed Hypergraphs
Directed hypergraphs are vital for modeling complex polyadic relationships in domains such as discrete mathematics, computer science, network security, and systems modeling. However, their inherent complexity often impedes effective visualization and analysis, particularly for large graphs. This paper introduces a novel Transitivity Preserving Projection (TPP) to address the limitations of the computationally intensive Basu and Blanning projection (BBP), which can paradoxically increase complex…

I finally figured out why one server had started barfing on 421 (no SNI) when most of the rest are similar and some have more domains and they're all behind haproxy with equivalent backend config.
This server has httpd-2.4.64-1 those that are ok are at httpd-2.4.62-1 with the other version available.
So I guess the others will fall down at the next update.
The fix is to send an explicit sni in the invocation in the particular back end section.
So I suppose I fi…

Unstable periodic orbits galore and quantum hyperscarring in highly frustrated magnets
Andrea Pizzi, Claudio Castelnovo, Johannes Knolle
https://arxiv.org/abs/2508.04763 https:/…

Unstable periodic orbits galore and quantum hyperscarring in highly frustrated magnets
Highly frustrated magnets, with their macroscopically-degenerate classical ground states and massively-entangled quantum spin liquid phases, have been pivotal to the development of modern condensed matter concepts such as emergent symmetries, topological order, and fractionalisation. The effects of frustration and massive degeneracies at high energy, where the many-body dynamics becomes chaotic, have hitherto been far less explored. Here, we identify a high-energy dynamical analog of highly-fru…

The Supermarket Model on a Dynamic Regular Hypergraph
John Fernley, Bal\'azs Gerencs\'er
https://arxiv.org/abs/2507.03931 https://

The Supermarket Model on a Dynamic Regular Hypergraph
The supermarket model is a system of $n$ queues each with serving rates $1$ and arrival rates $λ$ per vertex, where tasks will move on arrival to the shortest adjacent queue. We consider the supermarket model in the small $λ$ regime on a large dynamic configuration hypergraph with stubs swapping their hyperedge membership at rate $κ$. This interpolates previous investigations of the supermarket model on static graphs of bounded degree (where an exponential tail produces a logarithmic queue…

Single-Shot Decoding of Biased-Tailored Quantum LDPC Codes
Devon Campbell
https://arxiv.org/abs/2509.06316 https://arxiv.org/pdf/2509.06316

Single-Shot Decoding of Biased-Tailored Quantum LDPC Codes
Quantum processors are often affected by biased noise and noisy readout, which reduce reliability and reproducibility. This work combines two complementary strategies to address these challenges. The first is bias tailoring, which aligns stabilizers with the dominant error type. The second is single-shot (SS) decoding, which uses metachecks to identify measurement faults from just one noisy round. We implement these ideas in a four-dimensional lifted hypergraph product (4D-LHP) code constructed…

Brill--Noether loci in genus $\leq 12$
Richard Haburcak
https://arxiv.org/abs/2507.04902 https://arxiv.org/pdf/2507.04902

Brill--Noether loci in genus $\leq 12$
A refined Brill--Noether theory seeks to determine which linear series are admitted by a ``general'' curve in a particular Brill--Noether locus. However, as Brill--Noether loci are not irreducible in general, a coarse answer is given by the relative positions of Brill--Noether loci. Via an analysis of unstable Lazarsfeld--Mukai bundles on K3 surfaces, we distinguish Brill--Noether loci and provide expectations for the relative positions of Brill--Noether loci in general. Together with classical…

Mit mehreren Millionen km/h: Weiße Zwerge können einander aus Galaxie schleudern
Nachdem mehrere hyperschnelle Weiße Zwergsterne entdeckt wurden, gibt es nun eine Erklärung. Die Objekte sind das Produkt einer doppelten Sternenexplosion.

Realizing Scaling Laws in Recommender Systems: A Foundation-Expert Paradigm for Hyperscale Model Deployment
Dai Li, Kevin Course, Wei Li, Hongwei Li, Jie Hua, Yiqi Chen, Zhao Zhu, Rui Jian, Xuan Cao, Bi Xue, Yu Shi, Jing Qian, Kai Ren, Matt Ma, Qunshu Zhang, Rui Li
https://arxiv.org/abs/2508.02929 …

Realizing Scaling Laws in Recommender Systems: A Foundation-Expert Paradigm for Hyperscale Model Deployment
While scaling laws promise significant performance gains for recommender systems, efficiently deploying hyperscale models remains a major unsolved challenge. In contrast to fields where FMs are already widely adopted such as natural language processing and computer vision, progress in recommender systems is hindered by unique challenges including the need to learn from online streaming data under shifting data distributions, the need to adapt to different recommendation surfaces with a wide div…