Search for: All records

This paper describes an NSF (National Science Foundation) S-STEM-funded scholarship program, representing a collaborative five-year grant project among three prominent universities in the Southeast region of the United States. Its primary objective is to support dedicated scholars in graduating and finding a professional pathway. Each institution recruited a cohort of 15-20 scholars annually for three years. The project offers scholarships and provides curricular and cocurricular support to academically talented but financially challenged students in the computing disciplines, including Computer Science, Computer Engineering, Cybersecurity, and Information Technology majors, starting from their junior years. The program aims to impact 150 scholars, most of whom are underrepresented in computing. Scholars receive support throughout their graduation and beyond should they pursue graduate studies in a STEM (Science, Technology, Engineering, and Math) discipline at any of the three participating institutions. Besides funds, the program provides an expansive career pathway opportunity to each of its students, accompanied by various supporting services, a dedicated advising team, experiential learning offices, career services offices, and graduate schools. Supporting services include internship fairs, panel discussions with alumni, resume workshops, graduate school application workshops, and career fairs. The project brings together the unique collaboration of three institutions for each of its supported activities to significantly enhance the support and opportunities offered to its scholars and to conduct meaningful research studies that include significant-sized intersectional populations. 

The demand for computing professionals has grown exponentially due to the rapid expansion of technology and digitalization in various industries. As a result, understanding the importance of pathways into computing education and professions has become crucial. These pathways serve as structured routes that guide individuals in acquiring the necessary skills and knowledge to pursue careers in the computing field. Hence, it is essential for educational institutions to understand students’ perspectives, particularly those from lower-income socio-economic status, to broaden participation within computing education and professional fields. Though there are various pathways into computing education and professions, for the purposes of this research and the program, we review the existing literature about three primary pathways: graduate school, internship or industry profession, and entrepreneurship. 

The demand for computing professionals has grown exponentially due to the rapid expansion of technology and digitalization in various industries. As a result, understanding the importance of pathways into computing education and professions has become crucial. These pathways serve as structured routes that guide individuals in acquiring the necessary skills and knowledge to pursue careers in the computing field. Hence, it is essential for educational institutions to understand students’ perspectives, particularly those from lower-income socio-economic status, to broaden participation within computing education and professional fields. Though there are various pathways into computing education and professions, for the purposes of this research and the program, we review the existing literature about three primary pathways: graduate school, internship or industry profession, and entrepreneurship. 

The Florida IT Graduation Attainment Pathways (Flit-GAP), an NSF S-STEM, Track 3 grant effort, involves three public metropolitan institutions from Florida’s three most populous areas: Florida International University (FIU) in Miami, University of Central Florida (UCF) in Orlando, and University of South Florida (USF) in Tampa. Flit-GAP supports up to 50 students per year for each of the first 3 years of the project’; recruits are juniors from Computer Science, Information Technology, Computer Engineering, and Cybersecurity, and other computing majors. The relationship among the three institutions is formalized as the Consortium of Florida Metropolitan Research Universities. The consortium is a strategic priority of each institution. In Year 1, 42 students participated in the scholarship program at the three institutions (16 FIU; 14 UCF; 11 USF). 

The Florida IT Graduation Attainment Pathways (Flit-GAP), an NSF S-STEM, Track 3 grant effort, involves three public metropolitan institutions from Florida’s three most populous areas: Florida International University (FIU) in Miami, University of Central Florida (UCF) in Orlando, and University of South Florida (USF) in Tampa. Flit-GAP supports up to 50 students per year for each of the first 3 years of the project’; recruits are juniors from Computer Science, Information Technology, Computer Engineering, and Cybersecurity, and other computing majors. The relationship among the three institutions is formalized as the Consortium of Florida Metropolitan Research Universities. The consortium is a strategic priority of each institution. In Year 1, 42 students participated in the scholarship program at the three institutions (16 FIU; 14 UCF; 11 USF). 

The Florida IT Graduation Attainment Pathways (Flit-GAP), an NSF S-STEM, Track 3 grant effort, involves three public metropolitan institutions from Florida’s three most populous areas: Florida International University (FIU) in Miami, University of Central Florida (UCF) in Orlando, and University of South Florida(USF) in Tampa. Flit-GAP supports up to 50 students per year for each of the first 3 years of the project’; recruits are juniors from Computer Science, Information Technology, Computer Engineering, and Cybersecurity, and other computing majors. The relationship among the three institutions is formalized as the Consortium of Florida Metropolitan Research Universities. The consortium is a strategic priority of each institution. In Year 1, 42 students participated in the scholarship program at the three institutions (16 FIU; 14 UCF; 11 USF). 

Work on scaling laws has found that large language models (LMs) show predictable improvements to overall loss with increased scale (model size, training data, and compute). Here, we present evidence for the claim that LMs may show inverse scaling, or worse task performance with increased scale, e.g., due to flaws in the training objective and data. We present empirical evidence of inverse scaling on 11 datasets collected by running a public contest, the Inverse Scaling Prize, with a substantial prize pool. Through analysis of the datasets, along with other examples found in the literature, we identify four potential causes of inverse scaling: (i) preference to repeat memorized sequences over following in-context instructions, (ii) imitation of undesirable patterns in the training data, (iii) tasks containing an easy distractor task which LMs could focus on, rather than the harder real task, and (iv) correct but misleading few-shot demonstrations of the task. We release the winning datasets at https://inversescaling.com/data to allow for further investigation of inverse scaling. Our tasks have helped drive the discovery of U-shaped and inverted-U scaling trends, where an initial trend reverses, suggesting that scaling trends are less reliable at predicting the behavior of larger-scale models than previously understood. Overall, our results suggest that there are tasks for which increased model scale alone may not lead to progress, and that more careful thought needs to go into the data and objectives for training language models.