More Like this

1. Reply to Pielke (2025)

   https://doi.org/10.1175/JAMC-D-24-0236.1  

   Willoughby, H E; Hernandez, J I; Pinnock, Andrew (April 2025, Journal of Applied Meteorology and Climatology)

   Hu, Qi (Ed.)

   Abstract Pielke deprecates both the ICAT database, which he once recommended, and U.S. tropical cyclone (TC) damage estimates from the National Centers for Environmental Information (NCEI). We do not share these views. Willoughby et al. (hereafter WL24) is based upon ICAT damage for 1900–2017, both then-year and normalized for inflation, population, and individual wealth, extended to 2022 with National Hurricane Center (NHC) official figures from NCEI. Pielke represents the data of Weinkle et al. (hereafter WK18) as a superior source. We find troubling anomalies in the WK18 data. The issue is that WK18 find that normalized TC damage is constant, but WL24 find that it is increasing. Here, we replicate the WL24 analysis with WK18 data and find a statistically significant growth of then-year damage relative to the U.S. economy, a statistically significant increase in the occurrence of the most damaging TCs, and a 0.6% per year increase in TC normalized damage. The last of these is not statistically significant because of the large variance due to the modulation of TC impacts by the Atlantic multidecadal oscillation. Thus, the increase in U.S. TC damage is sufficiently robust to survive the shortcomings of both datasets. 

   more » « less
2. A Report on Financial Regulations Challenge at COLING 2025

   Wang, K; Patel, J; Shen, C; Kim, D; Zhu, A; Lin, A; Borella, L; Osborne, C; White, M; Yang, SY; et al (December 2024, arXiv preprints)

   Financial large language models (FinLLMs) have been applied to various tasks in business, finance, accounting, and auditing. Complex financial regulations and standards are critical to financial services, which LLMs must comply with. However, FinLLMs’ performance in understanding and interpreting financial regulations has rarely been studied. Therefore, we organize the Regulations Challenge, a shared task at COLING FinNLP-FNP-LLMFinLegal2025. It encourages the academic community to explore the strengths and limitations of popular LLMs. We create 9 novel tasks and corresponding question sets. In this paper, we provide an overview of these tasks and summarize participants’ approaches and results. We aim to raise awareness of FinLLMs’ professional capability in financial regulations. 

   more » « less
3. BLEND: A Unified Data Discovery System. In ICDE 2025.

   Esmailoghli, Mahdi; Schnell, Christoph; Miller, Renée; Abedjan, Ziawasch (May 2025, IEEE)

   Most research on data discovery has so far focused on improving individual discovery operators such as join, correlation, or union discovery. However, in practice, a combination of these techniques and their corresponding indexes may be necessary to support arbitrary discovery tasks. We propose BLEND, a comprehensive data discovery system that supports existing operators and enables their flexible pipelining. BLEND is based on a set of lower-level operators that serve as fundamental building blocks for more complex and sophisticated user tasks. To reduce the execution runtime of discovery pipelines, we propose a unified index structure and a rule- and cost-based optimizer that rewrites SQL statements into low-level operators when possible. We show the superior flexibility and efficiency of our system compared to ad-hoc discovery pipelines and stand-alone solutions. 

   more » « less
4. 2025 roadmap on 3D nanomagnetism

   https://doi.org/10.1088/1361-648X/ad9655  

   Gubbiotti, Gianluca; Barman, Anjan; Ladak, Sam; Bran, Cristina; Grundler, Dirk; Huth, Michael; Plank, Harald; Schmidt, Georg; van_Dijken, Sebastiaan; Streubel, Robert; et al (February 2025, Journal of Physics: Condensed Matter)

   Abstract The transition from planar to three-dimensional (3D) magnetic nanostructures represents a significant advancement in both fundamental research and practical applications, offering vast potential for next-generation technologies like ultrahigh-density storage, memory, logic, and neuromorphic computing. Despite being a relatively new field, the emergence of 3D nanomagnetism presents numerous opportunities for innovation, prompting the creation of a comprehensive roadmap by leading international researchers. This roadmap aims to facilitate collaboration and interdisciplinary dialogue to address challenges in materials science, physics, engineering, and computing. The roadmap comprises eighteen sections, roughly divided into three blocks. The first block explores the fundamentals of 3D nanomagnetism, focusing on recent trends in fabrication techniques and imaging methods crucial for understanding complex spin textures, curved surfaces, and small-scale interactions. Techniques such as two-photon lithography and focused electron beam-induced deposition enable the creation of intricate 3D architectures, while advanced imaging methods like electron holography and synchrotron x-ray tomography provide nanoscale spatial resolution for studying magnetization dynamics in three dimensions. Various 3D magnetic systems, including coupled multilayer systems, artificial spin-ice, magneto-plasmonic systems, topological spin textures, and molecular magnets are discussed. The second block introduces analytical and numerical methods for investigating 3D nanomagnetic structures and curvilinear systems, highlighting geometrically curved architectures, interconnected nanowire systems, and other complex geometries. Finite element methods are emphasized for capturing complex geometries, along with direct frequency domain solutions for addressing magnonic problems. The final block focuses on 3D magnonic crystals and networks, exploring their fundamental properties and potential applications in magnonic circuits, memory, and spintronics. Computational approaches using 3D nanomagnetic systems and complex topological textures in 3D spintronics are highlighted for their potential to enable faster and more energy-efficient computing. 

   more » « less
5. Model Lakes. In EDBT 2025.

   https://doi.org/10.48786/edbt.2025.81  

   Pal, Koyena; Bau, David; Miller, Renée J (January 2025, OpenProceedings.org)

   EDBT (Ed.)

   Given a set of deep learning models, it can be hard to find models appropriate to a task, understand the models, and characterize how models are different one from another. Currently, practi- tioners rely on manually-written documentation to understand and choose models. However, not all models have complete and reliable documentation. As the number of models increases, the challenges of finding, differentiating, and understanding mod- els become increasingly crucial. Inspired from research on data lakes, we introduce the concept of model lakes. We formalize key model lake tasks, including model attribution, versioning, search, and benchmarking, and discuss fundamental research challenges in the management of large models. We also explore what data management techniques can be brought to bear on the study of large model management. 

   more » « less