Tootfinder

The recording of today’s “Searching with AI” webinar for the Media Education Lab is now available!
https://mediaeducationlab.com/events/searching-ai
Our September meeting will move to Monday, Sept 8th because of the Labor Day U.S. holiday. It will be a “Favorite Podcast Share O…

Data Diversification Methods In Alignment Enhance Math Performance In LLMs
Berkan Dokmeci, Qingyang Wu, Ben Athiwaratkun, Ce Zhang, Shuaiwen Leon Song, James Zou
https://arxiv.org/abs/2507.02173

Data Diversification Methods In Alignment Enhance Math Performance In LLMs
While recent advances in preference learning have enhanced alignment in human feedback, mathematical reasoning remains a persistent challenge. We investigate how data diversification strategies in preference optimization can improve the mathematical reasoning abilities of large language models (LLMs). We evaluate three common data generation methods: temperature sampling, Chain-of-Thought prompting, and Monte Carlo Tree Search (MCTS), and introduce Diversified-ThinkSolve (DTS), a novel structur…

Neighbor-Sampling Based Momentum Stochastic Methods for Training Graph Neural Networks
Molly Noel, Gabriel Mancino-Ball, Yangyang Xu
https://arxiv.org/abs/2508.00267 https://

Neighbor-Sampling Based Momentum Stochastic Methods for Training Graph Neural Networks
Graph convolutional networks (GCNs) are a powerful tool for graph representation learning. Due to the recursive neighborhood aggregations employed by GCNs, efficient training methods suffer from a lack of theoretical guarantees or are missing important practical elements from modern deep learning algorithms, such as adaptivity and momentum. In this paper, we present several neighbor-sampling (NS) based Adam-type stochastic methods for solving a nonconvex GCN training problem. We utilize the con…

Holger #Zastrow wäre weniger Bürgerinnenbeteiligung beim Neubau der #Carolabrücke lieber gewesen.
Das kann man einfach mal so stehen lassen.

Schneller Carolabrückenbau: CDU, FDP und Team Zastrow stellen Änderungsantrag
Schneller Carolabrückenbau: CDU, FDP und Team Zastrow stellen Änderungsantrag | Wie soll die neue Carolabrücke aussehen? Für CDU, FDP/Freie Bürger und Team Zastrow ist die Antwort klar: leistungsfähig, schnell gebaut und mit vier Spuren …

Toch wel lekker dat je bij felde https://ardmediathek.de na een crimi een paar voorstellen voor een vervolg krijgt.
Wel jammer dat een aantal achter een regioblock zitten. Maar daar vallen gelukkig ook de nagesynchroniseerde Engelse series onder. 😀

Partial Weakly-Supervised Oriented Object Detection
Mingxin Liu, Peiyuan Zhang, Yuan Liu, Wei Zhang, Yue Zhou, Ning Liao, Ziyang Gong, Junwei Luo, Zhirui Wang, Yi Yu, Xue Yang
https://arxiv.org/abs/2507.02751

Partial Weakly-Supervised Oriented Object Detection
The growing demand for oriented object detection (OOD) across various domains has driven significant research in this area. However, the high cost of dataset annotation remains a major concern. Current mainstream OOD algorithms can be mainly categorized into three types: (1) fully supervised methods using complete oriented bounding box (OBB) annotations, (2) semi-supervised methods using partial OBB annotations, and (3) weakly supervised methods using weak annotations such as horizontal boxes o…

This https://arxiv.org/abs/2102.08600 has been replaced.
link: https://scholar.google.com/scholar?q=a

A new characterization of the convergence factor of two-level methods
Multilevel methods are among the most efficient numerical methods for solving large-scale systems of equations that arise from discretized partial differential equations. Two-level convergence theory plays a fundamental role in the analysis and design of multilevel methods. In this paper, we present a concise and easy-to-use identity for characterizing the convergence factor of two-level methods, whose hierarchical spaces can be either overlapping or non-overlapping. In order to illustrate its …

Graph-Embedding Empowered Entity Retrieval
Emma J. Gerritse, Faegheh Hasibi, Arjen P. de Vries
https://arxiv.org/abs/2506.03895 https://

Graph-Embedding Empowered Entity Retrieval
In this research, we investigate methods for entity retrieval using graph embeddings. While various methods have been proposed over the years, most utilize a single graph embedding and entity linking approach. This hinders our understanding of how different graph embedding and entity linking methods impact entity retrieval. To address this gap, we investigate the effects of three different categories of graph embedding techniques and five different entity linking methods. We perform a reranking…

#book2read: Thinking, Fast and Slow by Daniel Kahneman
#MediaEdClub

Symplectic Hamiltonian Hybridizable Discontinuous Galerkin Methods for Linearized Shallow Water Equations
C. N\'u\~nez (Facultad de Matem\'aticas, Pontificia Universidad Cat\'olica de Chile, Santiago, Chile), M. A. S\'anchez (Instituto de Ingenier\'ia Matem\'atica y Computacional, Facultad de Matem\'aticas y Escuela de Ingenier\'ia, Pontificia Universidad Cat\'olica de Chile, Santiago, Chile)

Symplectic Hamiltonian Hybridizable Discontinuous Galerkin Methods for Linearized Shallow Water Equations
This paper focuses on the numerical approximation of the linearized shallow water equations using hybridizable discontinuous Galerkin (HDG) methods, leveraging the Hamiltonian structure of the evolution system. First, we propose an equivalent formulation of the equations by introducing an auxiliary variable. Then, we discretize the space variables using HDG methods, resulting in a semi-discrete scheme that preserves a discrete version of the Hamiltonian structure. The use of an alternative form…