Pingry School

Despite advancements in quality of life and technology throughout the world, food insecurity has remained a challenge, consistently plaguing all nations. In the United States, modern food insecurity began during the Great Depression as the government introduced the first federal food program to moderate national economic hardship. Federal intervention in food insecurity soon became a hot topic for the next century as lawmakers debated federal policy shaping the landscape of food insecurity. This paper examines food insecurity through the dimensions that shape its current landscape: national and statewide politics, geography, and social inequality. Through this examination, this paper demonstrates that food insecurity presents a deep issue that can be categorized by a series of structural challenges, such as political polarization, geographic inequalities, and persistent racial and social disparities. This paper concludes by offering a solution to the modern problem of food insecurity by focusing on long-term solutions instead of following the contemporary model.

The capacity of computer-aided drug development to effectively explore vast chemical regions has led to an explosion in recent years. Reinforcement learning (RL), which learns through trial-and-error interactions, is uniquely suited to drug discovery and adaptive decision-making in Intelligent Environments (IEs). Three prominent reinforcement learning-based models for de novo molecular generation are examined in this review: REINVENT, a policy-based method that optimizes molecular design using an RNN and a prior network; MolDQN, a DQN-based method that presents molecular modification as a sequential decision process; and ORGANIC, a GAN-based model that learns to generate molecules by differentiating between high- and low-quality candidates. Comparing model architectures and key findings provide valuable information about the real-world uses and future direction of reinforcement learning in drug design. The adaptive and multi-modal nature RL lends itself to applications IE including personalized medicine and environmental toxicology.

The emergence of large-language models (LLMs) in everyday applications has highlighted their capability to provide human-like dialog and answer complex questions. However, LLMs often exhibit a tendency to "hallucinate," generating responses that appear factual but are incorrect, even when trained on ground truth data. In high-stakes applications such as medical diagnostics and mechanical troubleshooting, the need for accurate, trustworthy, and explainable AI systems is paramount. Without these qualities, such systems cannot be fully relied upon for critical decision-making. In this paper, we propose a novel AI system designed to address these challenges. The system integrates an AI agent, a logic model, an LLM, and a prompt engine to enhance accuracy and trustworthiness in a sequence of dialog. Experimental results demonstrate that the proposed system achieves an average accuracy of 86.49% across three diagnostic and troubleshooting tasks in healthcare, automotive maintenance, and digital equipment applications. In contrast, the baseline LLM only approach achieves an average accuracy of 63.42%, representing an absolute accuracy improvement of 23.07% with our approach. Additionally, the proposed system achieves a 4.26-fold speedup in response time across the experiments. These results highlight the potential of the proposed system to enhance reliability and efficiency in critical real-world applications.

Forecasting hurricane strength timely and accurately is of critical importance to ensure that preparations and evacuations can be carried out as necessary to save lives and property. In recent years, hurricanes have become more likely to experience rapid intensification, making them more dangerous and difficult to forecast with sufficient confidence, precision, and lead time. In general, rapid intensification is defined as a sudden increase in storm intensity within a short period of time, and often this intensification occurs just prior to landfall. Recent studies have linked the rapid intensification of storms to changes in the atmosphere and oceans resulting from climate change. The proportion of storms experiencing rapid intensification tripled from 1982 to 2009, suggesting that existing forecasting models based on historical storm data may be insufficient to account for today’s climate environment. Significant advances in satellite observation technology and an increase of in-situ storm data collected by the National Oceanic and Atmospheric Administration (NOAA) aircraft have greatly increased the information available for more accurate predictions of these storms. This paper will investigate several factors that may contribute to hurricane rapid intensification and assess their predictive value for Atlantic storms. A multivariable regression model was fit to the data to quantify predictive relationships and to provide a tool that can be used to help forecast rapid intensification of future storms. The analysis indicated a strong relationship between observed meteorological and physical storm data and hurricane rapid intensification, and the model produced an acceptable forecast success when applied to recent Atlantic storm history.

With the booming development of socio-economic and Internet technology, social culture has also continued to improve and enrich the times. Fan culture is given a new cultural connotation, and the fans cover the coverage of all sectors of society, which has attracted widespread attention. This study explores the correlation between parasocial relationships and celebrity worship among young Chinese women, utilizing questionnaires to examine how social connections influence celebrity worship behavior and its impact on well-being. The findings reveal a significant association between parasocial connections with celebrities and the level of celebrity worship. Despite potential adverse effects, celebrity worship provides emotional support and fosters a positive outlook on the future through parasocial relationships. The research also delves into the psychological mechanisms underlying the formation of these relationships, emphasizing the role of social media and celebrity culture. Understanding celebrity worship among young Chinese women contributes to our understanding of fan behavior, personal identity development, and the broader socio-cultural context of media and celebrity in contemporary society.

This chapter examines the exponential growth of the influencer industry in recent years, as well as the challenges associated with measuring the efficacy of influencer marketing and advertising in general. Influencer Marketing Hub defines influencers as people with the power to affect buying habits of consumers through social media, and influencer marketing as a brand working with one of these influencers either to market their products or services or simply just to improve brand reputation. Using data related to negative consumer impressions of traditional advertising features like pop-up ads and the untrustworthy review phenomenon and relating it to data about Generation Z’s modern opinions on social media marketing, this study explains how influencer marketing is the better choice to appeal to a younger audience. Three characteristics—physical beauty, social attractiveness, and attitude homophily—are reinforced in the development of pseudo-social interactions between the influencer and the customer. The main difference between influencer marketing and traditional marketing is the emotional and personal connection that the marketer, the influencer, can form with social media users.

In static advertisements, images instead of videos are used to promote products, and it is most often the faces of the models that get the consumers' attention. How facial expressions transfer emotions from the producer to the consumer has been explained by the emotional contagion theory, which suggests the recipient replicates the emitter through motor mimicry which changes the recipient's emotional state. This paper investigates the effects of facial expressions in advertisements on adolescents in comparison to adults. It was hypothesized that participants would give a higher product evaluation if the model displayed a real smile, versus a fake smile or neutral facial expression. It was also hypothesized that adolescents would be more susceptible to facial expressions and therefore pay more for the same product than adults. High school and adult participants completed a survey in which they evaluated a product (a white t-shirt) when the model displayed a neutral face, fake smile, and real smile. The results showed that participants gave higher product evaluation scores and were willing to pay more when the model displayed a genuine smile over the other facial expressions. In addition, adolescents were less influenced by the model's expressions than adults which may be due to their lack of experience in spending. To extend this study, the effects of emotional contagion theory on other factors including how the product attracts attention or lasts in the consumers' memory could be tested.

ABSTRACT Undergraduate STEM students majoring in various science sub-disciplines (e.g. chemistry, physics, engineering) must develop strong understandings of core foundational thermodynamics concepts. The ability for course instructors and researchers to effectively refine instruction and develop interventions to support students’ learning hinges on their ability to accurately gauge students’ knowledge through the use of established measures. The Thermodynamics Conceptual Reasoning Inventory (TCRI) is designed to gauge undergraduate students’ understanding of introductory thermodynamics concepts. The present study extends the findings of a previous publication by positioning the TCRI within the broader international literature of thermodynamics concept inventories and generating an argument for the reliability and validity of TCRI scores in a broader context. Participants (n = 278) took the revised 36-item TCRI (available in the supplementary online materials). Findings revealed that TCRI scores are useful in the broader context (e.g. no evidence of floor or ceiling effects, evidence of high reliability, no differences for students across majors, and TCRI scores were moderately correlated with both course exam scores and GPA). No further revisions are recommended based on analysis of item properties. The cumulative body of evidence related to the TCRI suggests that scores are useful indicators of undergraduate students’ conceptual understanding of introductory thermodynamics concepts.

Individuals are asked to make high‐impact decisions that become the world we live in, informing the treatment of addiction, gun violence, bioethics laws, and global climate change. Decisions of this magnitude should be made with evidence from primary source research. Yet few methods effectively teach this skill to high school students. Here we describe a highly coordinated system of 7 teams of students that train others with little scientific training or presentation skill to present research topics that are relevant and controversial. These presenters have little to no scientific or presentation training. The teams are capable of training one presenter a week over the course of a year. The system is self‐propagating and can be maintained with minimal faculty involvement. Our data indicate that a single faculty member can engage a team of 12 that in turn engages a student body of 30–70 students weekly. Further details will be shared at www.project80.orgThis abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.