Tootfinder

Irsay-Gordon on her hands-on approach: 'I would suggest it for anyone that has to pay coaches millions' https://www.espn.com/nfl/story/_/id/45501891/nfl-indianapolis-colts-ownership-carlie-irsay-gordon

A profile of Pat McAfee, among the top American sports pundits with millions of followers and an ESPN show, and a sports bro culture dominated by ex-NFL players (Aaron Timms/The Guardian)
https://www.theguardian.com/sport/ng-interactive/2025/…

Man attempts high jump at Trackapalooza 2025 held at Groton High School way in the Northeast corner of Tompkins County
#photo #photography #sports

MachineLearningLM: Continued Pretraining Language Models on Millions of Synthetic Tabular Prediction Tasks Scales In-Context ML
Haoyu Dong, Pengkun Zhang, Mingzhe Lu, Yanzhen Shen, Guolin Ke
https://arxiv.org/abs/2509.06806

MachineLearningLM: Continued Pretraining Language Models on Millions of Synthetic Tabular Prediction Tasks Scales In-Context ML
Large language models (LLMs) possess broad world knowledge and strong general-purpose reasoning ability, yet they struggle to learn from many in-context examples on standard machine learning (ML) tasks, that is, to leverage many-shot demonstrations purely via in-context learning (ICL) without gradient descent. We introduce MachineLearningLM, a portable continued-pretraining framework that equips a general-purpose LLM with robust in-context ML capability while preserving its general knowledge an…

Investigation: Israel's Unit 8200 built a system to collect millions of mobile phone calls made daily in Gaza and the West Bank using Microsoft's Azure platform (The Guardian)
https://www.theguardian.com/world/2025/aug/06/microso…

How millions of sugar cubes stopped polio
https://www.youtube.com/watch?v=vpM7bVVIMxk

Beyond Communication Overhead: A Multilevel Monte Carlo Approach for Mitigating Compression Bias in Distributed Learning
Ze'ev Zukerman, Bassel Hamoud, Kfir Y. Levy
https://arxiv.org/abs/2507.05508

Beyond Communication Overhead: A Multilevel Monte Carlo Approach for Mitigating Compression Bias in Distributed Learning
Distributed learning methods have gained substantial momentum in recent years, with communication overhead often emerging as a critical bottleneck. Gradient compression techniques alleviate communication costs but involve an inherent trade-off between the empirical efficiency of biased compressors and the theoretical guarantees of unbiased compressors. In this work, we introduce a novel Multilevel Monte Carlo (MLMC) compression scheme that leverages biased compressors to construct statistically…

Three days later, I've finished turning 20-odd pounds of whole frozen tomatoes (almost all of last summer's harvest from a friend's garden) into just over 3 litres of sauce. A bit of an ordeal (thawing, chopping, roasting, food-milling, reducing, reducing, reducing) but it tastes good!
Next time, fresh plum-style 'maters for sure!

Subgrid BoostCNN: Efficient Boosting of Convolutional Networks via Gradient-Guided Feature Selection
Biyi Fang, Jean Utke, Truong Vo, Diego Klabjan
https://arxiv.org/abs/2507.22842

Subgrid BoostCNN: Efficient Boosting of Convolutional Networks via Gradient-Guided Feature Selection
Convolutional Neural Networks (CNNs) have achieved remarkable success across a wide range of machine learning tasks by leveraging hierarchical feature learning through deep architectures. However, the large number of layers and millions of parameters often make CNNs computationally expensive to train, requiring extensive time and manual tuning to discover optimal architectures. In this paper, we introduce a novel framework for boosting CNN performance that integrates dynamic feature selection w…

The Living Library of Trees: Mapping Knowledge Ecology in the Arnold Arboretum
Johan Malmstedt, Giacomo Nanni, Dario Rodighiero
https://arxiv.org/abs/2509.00114 https://

The Living Library of Trees: Mapping Knowledge Ecology in the Arnold Arboretum
As biodiversity loss and climate change accelerate, botanical gardens serve as vital infrastructures for research, education, and conservation. This project focuses on the Arnold Arboretum of Harvard University, a 281-acre living museum founded in 1872 in Boston. Drawing on more than a century of curatorial data, the research combines historical analysis with computational methods to visualize the biographies of plants and people. The resulting platform reveals patterns of care and scientific o…