Tootfinder

Building High-Quality Datasets for Portuguese LLMs: From Common Crawl Snapshots to Industrial-Grade Corpora
Thales Sales Almeida, Rodrigo Nogueira, Helio Pedrini
https://arxiv.org/abs/2509.08824

Building High-Quality Datasets for Portuguese LLMs: From Common Crawl Snapshots to Industrial-Grade Corpora
The performance of large language models (LLMs) is deeply influenced by the quality and composition of their training data. While much of the existing work has centered on English, there remains a gap in understanding how to construct effective training corpora for other languages. We explore scalable methods for building web-based corpora for LLMs. We apply them to build a new 120B token corpus in Portuguese that achieves competitive results to an industrial-grade corpus. Using a continual pre…

OmniEVA: Embodied Versatile Planner via Task-Adaptive 3D-Grounded and Embodiment-aware Reasoning
Yuecheng Liu, Dafeng Chi, Shiguang Wu, Zhanguang Zhang, Yuzheng Zhuang, Bowen Yang, He Zhu, Lingfeng Zhang, Pengwei Xie, David Gamaliel Arcos Bravo, Yingxue Zhang, Jianye Hao, Xingyue Quan
https://arxiv.org/abs/2509.09332

OmniEVA: Embodied Versatile Planner via Task-Adaptive 3D-Grounded and Embodiment-aware Reasoning
Recent advances in multimodal large language models (MLLMs) have opened new opportunities for embodied intelligence, enabling multimodal understanding, reasoning, and interaction, as well as continuous spatial decision-making. Nevertheless, current MLLM-based embodied systems face two critical limitations. First, Geometric Adaptability Gap: models trained solely on 2D inputs or with hard-coded 3D geometry injection suffer from either insufficient spatial information or restricted 2D generalizat…

Isometry groups of Polish ultrametric spaces
Riccardo Camerlo, Alberto Marcone, Luca Motto Ros
https://arxiv.org/abs/2508.08480 https://arxiv.org/pdf/2508.…

Isometry groups of Polish ultrametric spaces
We solve a long-standing open problem, formulated by Krasner in the 1950's, in the context of Polish (i.e. separable complete) ultrametric spaces by providing a characterization of their isometry groups using suitable forms of generalized wreath products of full permutation groups. Since our solution is developed in the finer context of topological (Polish) groups, it also solves a problem of Gao and Kechris from 2003. Furthermore, we provide an exact correspondence between the isometry groups …

"de gemeente Emmen [heeft] alle procedures versneld, vergunningen verleend en grond beschikbaar gesteld om de bouw van het station mogelijk te maken.
[..]
Tennet, dat de vertraging deels zou verklaren doordat de bouwvergunning eerder dan verwacht onherroepelijk werd. "Een meevaller in de juridische procedures wordt nu een excuus om drie tot vijf jaar uitloop te accepteren," "
Ik snap de frustratie goed.

Emmer raad haalt hard uit naar Tennet: 'Vertraging hoogspanningsstation Veenoord is onacceptabel'
"Een meevaller in de procedures wordt nu een excuus om drie tot vijf jaar uitloop te accepteren. Dat is geen overmacht, dat is een verkeerde inschatting en tekortschietend risicomanagement. En dat mag niet op Emmen worden afgewenteld."

Ready to record a new episode of my ✨HACKER PODCAST✨

New York-based Crosby, a law firm that uses AI to help speed up the contracts process, raised a $20M Series A co-led by Index Ventures, BCV, and Elad Gil (Alex Konrad/Upstarts Media)
https://www.upstartsmedia.com/p/crosby-ai-law-firm-raises-20-million

Crosby, 'AI Law Firm' Working With Startups Like Cursor, Raises $20 Million
The Series A from Index Ventures and Bain Capital Ventures will help lawyer turned founder Ryan Daniels expand how his year-old startup can use AI to speed up contract reviews.

python_dependency: Python Dependency Network
Python's package dependency networks. Nodes in the network are Python's packages registered to PyPI and edges are dependencies among packages.
This network has 58743 nodes and 108399 edges.
Tags: Technological, Software, Unweighted
https://networks.sk…

Emission of Nambu-Goldstone bosons from the semilocal string network
Yukihiro Kanda, Naoya Kitajima
https://arxiv.org/abs/2510.07894 https://arxiv.org/pdf/…

Emission of Nambu-Goldstone bosons from the semilocal string network
Semilocal cosmic string is a line-like non-topological soliton associated with the breakdown of the $SU(2)_{\rm global} \times U(1)_{\rm gauge}$ symmetry to the $U(1)_{\rm global}$ symmetry. The broken phase has two massless Nambu-Goldstone (NG) modes as dynamical fields, and they can be emitted by semilocal strings. In this paper, we numerically show that such NG bosons are copiously produced with the evolution of the semilocal string network in the early universe. Our numerical analysis shows…

First Lego table leaf is built and attached. Mistakes were made (and corrected). It looks ok. The gap (1/4") is a little more than I wanted, but I'll know how to fix that going forward (measure better and build over sized). I am happy that it stands up with just one leaf attached. I was concerned that not having the counter balance of the other leaf would cause it to tip.
#Woodworking

Parallel-in-Time Solution of Allen-Cahn Equations by Integrating Operator Learning into the Parareal Method
Yuwei Geng, Junqi Yin, Eric C. Cyr, Guannan Zhang, Lili Ju
https://arxiv.org/abs/2510.07672

Parallel-in-Time Solution of Allen-Cahn Equations by Integrating Operator Learning into the Parareal Method
While recent advances in deep learning have shown promising efficiency gains in solving time- dependent partial differential equations (PDEs), matching the accuracy of conventional numerical solvers still remains a challenge. One strategy to improve the accuracy of deep learning-based solutions for time- dependent PDEs is to use the learned solution as the coarse propagator in the Parareal method and a traditional numerical method as the fine solver. However, successful integration of deep lear…