DARWIN Series: Domain Specific Large Language Models for Natural Science [PDF]
arXiv.org, 2023 Emerging tools bring forth fresh approaches to work, and the field of natural science is no different. In natural science, traditional manual, serial, and labour-intensive work is being augmented by automated, parallel, and iterative processes driven by ...
Tong Xie +11 more
semanticscholar +1 more source
DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation [PDF]
International Conference on Machine Learning, 2022 We introduce DS-1000, a code generation benchmark with a thousand data science problems spanning seven Python libraries, such as NumPy and Pandas. Compared to prior works, DS-1000 incorporates three core features.
Yuhang Lai +9 more
semanticscholar +1 more source
Development of natural science through the Gamification and ICT in Primary Education
Edutec, 2022 Innovative methodologies and new emerging technologies have significant potential for education. Therefore, their use is becoming more and more common in the educational society.
Eider Bilbao-Aiastui +1 more
doaj +1 more source
Developing Interactive Multimedia for Natural Science in High School
International Journal of Multidisciplinary Research of Higher Education, 2023 This study aims to create interactive multimedia products in learning Natural Sciences, as well as to analyze how the effectiveness of the products produced. This research uses the method Research and Development (R&D) model Borg & Gall. The research was
M. Iskandar +5 more
semanticscholar +1 more source
Nature, Energy and Society—A Scientific Study of the Options Facing Civilisation Today
Journal of Nuclear Engineering, 2022 The nations of the world plan to stop burning carbon fuels but have not fixed on any replacement. For social and economic confidence, they need to share a proper picture of the options.
Wade Allison
doaj +1 more source
Event Recognition Based on Deep Learning in Chinese Texts. [PDF]
PLoS ONE, 2016 Event recognition is the most fundamental and critical task in event-based natural language processing systems. Existing event recognition methods based on rules and shallow neural networks have certain limitations. For example, extracting features using
Yajun Zhang +3 more
doaj +1 more source
Conceptual Playworld as a Method of Facilitating Learning Beyond Subject Matter in Elementary School
Nordisk tidsskrift for utdanning og praksis, 2023 Children experience less time to play due to the focus on learning in early childhood education and earlier school starting times. Paradoxically, play is important for children’s learning and contributes to increased curiosity, wonder, and learning ...
Anne-Line Bjerknes +2 more
doaj +1 more source
NATURAL SCIENCE EDUCATION IN PRIMARY SCHOOL: SOME SIGNIFICANT POINTS
Journal of Baltic Science Education, 2022 The problems of natural science education remain extremely relevant. Over the last two decades, international conferences and symposia dedicated to the study of issues of natural science education have been held in many countries and at various levels ...
Vincentas Lamanauskas
semanticscholar +1 more source
Quantitative change of meta-cognitive activities of students effected by problem-solving teaching [PDF]
Zbornik Instituta za pedagoška istraživanja, 2006 Based on Fleivel’s theoretical framework, the subject of this paper is the study of changes in frequency of meta-cognitive activities affected by a two-month implementation of problem solving teaching in the subject of natural studies.
Miščević Gordana
doaj +1 more source
Die Frage nach der adäquaten Naturerkenntnis bei Immanuel Hermann Fichte
Internationale Zeitschrift für Kulturkomparatistik, 2022 The present article addresses the question of the adequate knowledge of nature in the context of Immanuel Hermann Fichte’s philosophy of nature. After an examination of the position and role that this systematic problem has both in Karl Joël’s book Der ...
Cristián Hernández Maturana
doaj +1 more source