71446 research outputs found
Sort by
Evaluation of Architectural Synthesis Using Generative AI: A case study on Palladio’s architecture
No full textRecent advancements in multimodal Generative AI may democratize specialized architectural tasks like interpreting technical drawings and creating 3D CAD models which traditionally require expert knowledge. This paper presents a comparative evaluation study of two systems—GPT-4o and Claude 3.5—in the task of architectural 3D synthesis. It takes as a case study two buildings in Palladio’s Four Books of Architecture (1965): Villa Rotonda and Palazzo Porto. High-level architectural models and drawings of the buildings were prepared inspired by Palladio’s original text and drawing corpus. Through sequential text and image prompting, the study characterizes intrinsic abilities of the systems in (1) interpreting 2D/3D representations of buildings from drawings, (2) encoding the buildings into a CAD software script, and (3) self-improving based on outputs. While both systems successfully generate individual parts, they struggle to accurately assemble these parts into the desired spatial relationships, with Claude 3.5 showing overall better performance, especially in self-correcting its output. The study contributes to ongoing research on benchmarking the strengths and weaknesses of off-the-shelf AI systems in intelligent human tasks requiring discipline-specific knowledge. The results show the potential of language-enabled AI systems to act as collaborative technical assistants in the architectural design process.Applied ComputationAccepted Manuscrip
Altitude Sickness: Comparing the Political Economy of Two Periods of Extreme Dollar Strength
No full textThis thesis examines the absence of coordinated international intervention to counter the US dollar’s historic strength in the 2020s, contrasting this with the 1980s episode that culminated in the Plaza Accord.
The study employs econometric analysis, historical case studies, and policy research to examine changes in global responses to dollar appreciation, with a focus on structural, cyclical, and geopolitical factors. Structurally, the decline of US manufacturing’s economic and political influence, driven by globalization and a shift toward high-value-added, knowledge-intensive industries (e.g., pharmaceuticals, advanced technology, and product design), along with the shrinkage of low-value-added, labor-intensive sectors such as textiles, has reduced sectoral vulnerability to exchange rate fluctuations.
Concurrently, the rise of offshore production and foreign direct investment in critical sectors like automotive manufacturing has further diluted domestic opposition to dollar strength. Cyclically, record postwar corporate profitability and a resilient labor market, characterized by an unemployment rate near 4 percent, and post-COVID-19 labor scarcity, stand in stark contrast to the 1980s, when industrial bankruptcies and double-digit joblessness amplified demands for currency intervention. Geopolitically, reduced systemic risks owing to emerging markets’ improved debt management and greater reliance on local currency financing, the euro’s stabilization of intra-European exchange rates, and China’s strategic aversion to replicating Japan’s Plaza-driven decline have diminished incentives for collective action.
The dollar’s rise reflects fundamentals, including growth differentials, energy self-reliance, non-price export competitiveness, and interest rate advantages rather than speculative excess.
Meanwhile, inflation-targeting regimes and the Federal Reserve’s domestic mandate preclude monetary easing for currency goals, while US-China strategic rivalry and G20 fragmentation render multilateral coordination implausible.
The analysis concludes that absent 1980s-style vulnerabilities, efforts to weaken the dollar lack economic justification or political traction, signaling a paradigm shift from multilateral currency diplomacy to unilateralism in an era of structural decoupling and great-power competition. These findings redefine the role of exchange rates in a multipolar world and challenge assumptions about the feasibility of international economic governance.Extension Studie
Multivalent Evaluation of Digital Preservation Success
No full textPurpose – Current assessment practice for digital preservation focuses on instrumental trustworthiness of processes, systems, and outputs, with insufficient regard for teleological success of resulting programmatic outcomes. This paper provides a communicological formalization of digital preservation success.
Design/methodology/approach – Digital preservation is repositioned as semiotic sense-making unfolding across archival timespans and accompanying cultural as well as technical distance, and not just as persistent data management.
Findings – Inherent intersubjective contingency of any use of preserved information objects necessitates multivalent consideration of various communicative qualities, experiential affordances, and artifactual characteristics for comprehensive determination of preservation success.
Research limitations/implications – The proposed conceptual framework provides a foundation for subsequent development of evaluative best practice recommendations.
Originality/value – While digital preservation concerns span a socio-technic continuum, evaluative practice should properly emphasize the embodied experience of preservation-enabled communication. Communicological framing of preservation assessment complements the predictive reliability of programmatic trustworthiness with confirmatory assurance of experiential efficacy.Accepted Manuscrip
Building a Science of Teaching Reading and Vocabulary: Experimental Effects of Structured Supplements for a Read Aloud Lesson on Third Graders’ Domain-Specific Reading Comprehension
No full textPurpose: This study contributes to the science of teaching reading and vocabulary by illustrating how a ubiquitous classroom practice – read alouds – can be enhanced by structured supplements. This experimental study examines whether and to what extent providing structured supplements can improve student comprehension outcomes by helping teachers foster discussions about academic vocabulary that support schema transfer as students make connections between known and new topics.
Method: A total of 80 third-grade teachers and their students (N = 965; 32% Black, 31% Hispanic, 25% white, 9 Asian, 48% Male) were randomly assigned to treatment or control conditions. Treatment students received a single social studies read aloud on the story of Apollo 11 with structured supplements while control students received the same read aloud story but without structured supplements. Students were from linguistically, economically, and ethnically diverse backgrounds.
Results: Effect sizes from hierarchical linear models indicated that students in the treatment condition outperformed students in the control condition on four measures of domain-specific reading comprehension: recall (ES=.17), near-transfer (ES=.17), mid-transfer (ES=.18), and content comprehension (ES=.18). Further exploratory analyses using structural equation modeling revealed that teacher language scaffolds—that is, temporary dialogic supports in which teachers went above and beyond the intervention script—explained 66% of the treatment effect on domain-specific reading comprehension.
Conclusion: Results from this study suggest that read alouds, when enhanced with structured supplements designed to facilitate schema transfer, can increase the amount of academic vocabulary teachers use during classroom instruction and improve their students’ ability to comprehend disciplinary texts.Accepted Manuscrip
AP® STEM Student Assessment of ChatGPT Prompt Responses
No full textThis thesis explores the current accuracy of generative AI ChatGPT responses to Advanced Placement® (AP®) free response and multiple-choice questions found in high school courses covering AP® Physics 1 and AP® Calculus AB. Moreover, this thesis compares how high school students evaluate ChatGPT responses to algebra, precalculus, and calculus concepts found within these high school courses at participating high schools in the DFW area of Texas. In general, studies on generative AI use within STEM high school classrooms across the United States is limited. Current literature suggests that generative AI programs like ChatGPT have the potential to supplement classroom instruction by providing personalized assistance and immediate access to subject-specific information; however, there is a noticeable gap in understanding how high school AP® students perceive and interact with AI-generated responses, particularly in relation to accuracy and effectiveness as a study aid. Based on a literature review undertaken as a part of this thesis, a mixed-methods study was developed focusing on the perceptions of high school students interacting with ChatGPT written responses to AP®-style free response questions. The results of this thesis study show that while ChatGPT can offer detailed explanations and improve student understanding of complex topics, its generated responses can lack mathematical accuracy. Students generally viewed ChatGPT as a useful educational resource but often struggled to distinguish between AI-generated responses and official solutions provided by College Board®. Lower-performing students were more likely to overestimate the accuracy and completeness of ChatGPT’s outputs, potentially due to limited subject matter understanding. This study highlights the importance of developing high school students’ critical evaluation skills and suggests that integrating educational AI like ChatGPT into the classroom requires careful consideration of generative AI limitations and potential impact on learning outcomes
'Will they be assets or will they be liabilities?': Two German Jewish Pediatrician Refugees from Nazi Germany and their Contributions to US Science and Society
No full textAuthor's Origina
Healing and Harming: Patients, Doctors, and Trust in Post-Independence India
No full textVersion of Recor
High continuity of forager ancestry in the Neolithic of the eastern Maghreb
No full textAncient DNA from the Mediterranean region has revealed long-range connections and population transformations associated with the spread of food producing economies [1-6]. However, in contrast to Europe, genetic data from this key transition in northern Africa are limited, and have only been available from the far western Maghreb (Morocco) [1-3]. Here, we present genome-wide data for nine individuals from the Later Stone Age (LSA) through the Neolithic in Algeria and Tunisia. The earliest individuals cluster with pre-Neolithic people of the western Maghreb (~15000-7600 Before Present (BP)), showing that this “Maghrebi” ancestry profile had a substantial geographic and temporal extent. At least one individual from Djebba (Tunisia), dating to ~8000 BP, harbored ancestry from European hunter-gatherers, likely reflecting early Holocene movement across the Strait of Sicily. Later Neolithic people from the eastern Maghreb retained largely local forager ancestry together with smaller contributions from European farmers (by ~7000 BP) and Levantine groups (by ~6800 BP), and were thus far less impacted by external gene flow than were populations in other parts of the Neolithic Mediterranean.Human Evolutionary BiologyAccepted Manuscrip