109812 research outputs found
Sort by
The Devil In The Bakken
No full textIn 2008 the oil boom hit North Dakota. The oil brought wealth and human health concerns to the state and specifically, the Fort Berthold Indian Reservation. In July of 2024 the largest Clean Air Act Settlement of its kind happened on the reservation due to Marathon Oil emitting thousands of tons more pollutants that regulations outlined. This story dives into the human health effects, advocacy and the difficulty in balancing health and wealth on the reservation
Prompt-Tuned Multi-Task Taxonomic Transformer (PTMTTaxoFormer)
No full textHierarchical Text Classification (HTC) is a subclass of multi-label classification, it is challenging because the hierarchy typically has a large number of diverse topics. Existing methods for HTC fall within two categories, local methods (a classifier for each level, node, or parent) or global methods (a single classifier for everything). Local methods are computationally expensive, whereas global methods often require complex explicit injection of the hierarchy, verbalizers, and/or prompt engineering. In this work, we propose Prompt Tuned Multi Task Taxonomic Transformer, a single classifier that uses a multi-task objective to predict one or more topics. The approach is capable of understanding the hierarchy during training without explicit injection, complex heads, verbalizers, or prompt engineering. PTMTTaxoFormer is a novel model architecture and training paradigm using differentiable prompts and labels that are learnt through backpropagation. PTMTTaxoFormer achieves state of the art results on several HTC benchmarks that span a range of topics consistently. Compared to most other HTC models, it has a simpler yet effective architecture, making it more production-friendly in terms of latency requirements (a factor of 2-5 lower latency). It is also robust and label-efficient, outperforming other models with 15\%-50\% less training data
Part 1: Mathematical Alchemy
Full text linkIn this set of four articles, we invite Calculus teachers to encounter some of the ideas that shaped and motivated the innovation of the subject, presented in story form. We find that in traditional treatments of these topics, the dramatic tension between ways of thinking, representations, and phenomena in the world can be lost in a sea of calculations, procedures, and partially-understood formal proofs. However, the fundamental ideas and insights can be identified in nascent form in students’ ways-of-thinking about Calculus, and in their struggles to connect Calculus with their prior mathematical experiences.
We are honored to contribute these Stories of Calculus to this number of The Mathematics Enthusiast, honoring the life and thinking of David Tall. David’s work with technology in Calculus learning brought him into close contact with Jim Kaput, the SimCalc Projects, and the Kaput Center. He and Jim discussed ideas that came to represent his “three worlds” of mathematics, a story he tells in part in his contribution to the volume, The SimCalc Vision and Contributions (Hegedus & Roschelle, 2013). His ideas, and his sense of both the challenges of learning Calculus and the potential transformations offered by digital and executable representations, are coherent with the approach we are advocating in these articles
Montana Kaimin, February 13, 2025
Full text linkStudent newspaper of the University of Montana, Missoula.https://scholarworks.umt.edu/studentnewspaper/11165/thumbnail.jp