Results 41 to 50 of about 118 (62)

Reasoning Vectors: Transferring Chain-of-Thought Capabilities via Task Arithmetic [PDF]

open access: yes
Large language models often require costly optimization, such as reinforcement learning, to master complex reasoning tasks. This work demonstrates that reasoning ability, once learned, can be extracted and transferred between models as a compact task ...
Zbeeb, Mohammad   +2 more
core  

SciDFM: A Large Language Model with Mixture-of-Experts for Science [PDF]

open access: yes
Recently, there has been a significant upsurge of interest in leveraging large language models (LLMs) to assist scientific discovery. However, most LLMs only focus on general science, while they lack domain-specific knowledge, such as chemical molecules ...
Shen, Zhennan   +9 more
core  

INVESTIGATION OF ARTIFICIAL INTELLIGENCE SUPPORTED AUTOMATIC ASSESSMENT AND EVALUATION SYSTEMS IN SCIENCE EDUCATION IN TERMS OF QUALITY

open access: yes
This study examines the quality dimensions of artificial intelligence-basedautomated assessment systems in science education in terms of validity,reliability, and conceptual understanding.
Özen, Gül İrem, Selvi, Mahmut
core  

Improving Weak-to-Strong Generalization with Scalable Oversight and Ensemble Learning [PDF]

open access: yes
This paper presents a follow-up study to OpenAI's recent superalignment work on Weak-to-Strong Generalization (W2SG). Superalignment focuses on ensuring that high-level AI systems remain consistent with human values and intentions when dealing with ...
Zhang, Jing   +7 more
core  

Layer Importance and Hallucination Analysis in Large Language Models via Enhanced Activation Variance-Sparsity [PDF]

open access: yes
Evaluating the importance of different layers in large language models (LLMs) is crucial for optimizing model performance and interpretability. This paper first explores layer importance using the Activation Variance-Sparsity Score (AVSS), which combines
Song, Zichen   +3 more
core  

Extracting the Wisdom of Crowds From Crowdsourcing Platforms

open access: yes, 2019
Enabled by the wave of online crowdsourcing activities, extracting the Wisdom of Crowds (WoC) has become an emerging research area, one that is used to aggregate judgments, opinions, or predictions from a large group of individuals for improved decision ...
Du, Qianzhou
core  

“The Growth Is There–But It’s Not That Evident, Is It!”: A Study in Science Delivery Improvement

open access: yesJournal of Science Teacher Education, 2005
The purpose of this study was to monitor the development of teacher personal attribute and school environmental factors at an urban intermediate (years 7 and 8) school in New Zealand as a result of an extended science professional development ...
Brian Lewthwaite
exaly   +4 more sources

The Application of a Primary Science Delivery Evaluation Questionnaire [PDF]

open access: yesResearch in Science Education, 2004
The purpose of this investigation was to apply a validated curriculum evaluation instrument, the Science Curriculum Implementation Questionnaire (SCIQ), in an educational context in which a science curriculum review had recently been completed and, by so
Brian Lewthwaite, Darrell Fisher
exaly   +2 more sources

T-SciQ: Teaching Multimodal Chain-of-Thought Reasoning via Large Language Model Signals for Science Question Answering

open access: yesProceedings of the AAAI Conference on Artificial Intelligence
Large Language Models (LLMs) have recently demonstrated exceptional performance in various Natural Language Processing (NLP) tasks. They have also shown the ability to perform chain-of-thought (CoT) reasoning to solve complex problems. Recent studies have explored CoT reasoning in complex multimodal scenarios, such as the science question answering ...
Heng Tao Shen
exaly   +5 more sources

Home - About - Disclaimer - Privacy