Search for: All records

The development and measurable improvements in performance of large language models on natural language tasks opens the opportunity to utilize large language models in an educational setting to replicate human tutoring, which is often costly and inaccessible. We are particularly interested in large language models from the GPT series, created by OpenAI. In the original study we found that the quality of explanations generated with GPT-3.5 was poor, where two different approaches to generating explanations resulted in a 43% and 10% successrate. In a replication study, we were interested in whether the measurable improvements in GPT-4 performance led to a higher rate of success for generating valid explanations compared to GPT-3.5. A replication of the original study was conducted by using GPT-4 to generate explanations for the same problems given to GPT-3.5. Using GPT-4, explanation correctness dramatically improved to a success rate of 94%. We were further interested in evaluating if GPT-4 explanations were positively perceived compared to human-written explanations. A preregistered, follow-up study was implemented where 10 evaluators were asked to rate the quality of randomized GPT-4 and teacher-created explanations. Even with 4% of problems containing some amount of incorrect content, GPT-4 explanations were preferred over human explanations. 

Seismic ground motion creates low-frequency atmospheric sound (infrasound) that is detectable at remote sensor arrays. However, earthquake infrasound signal analysis is complicated by interference between multiple waves arriving at sensors simultaneously, reducing the accuracy and detail of ground motion detection. Here we show that individual waves in complicated wavefields can be resolved by recording infrasound on large-N arrays and processing with CLEAN beamforming. Examining both a local (ML3.5, purely tropospheric infrasound propagation) and regional earthquake (ML6.5, upper-atmospheric returns), we detect infrasound from tens of km away and up to several hundred km away respectively. Source regions span arcs of approximately 90°, indicating that although detection bias does occur (most likely from atmospheric winds) the recorded infrasound sources are widely dispersed and not simply epicentral. Infrasound-based remote detection of ground motion over wide areas can complement point measurements by seismometers and spur innovations in earthquake research and real-time hazard monitoring.

 

This article outlines the key components of the River’s Edge Construction lesson plan. An explanation of how the lesson was delivered is presented alongside suggestions for implementation by K–6 teachers. The integration of scientific literacy is discussed first, followed by a discussion of each of the 5Es (Bybee et al. 2006). A timeframe for distributing the lesson phases is given; however, the activities included in this plan (see Supplementary Resources for specific lesson materials), should be modified to meet the needs and interest of students, and to align with allotted instructional time and objectives. 

This article outlines the key components of the River’s Edge Construction lesson plan. An explanation of how the lesson was delivered is presented alongside suggestions for implementation by K–6 teachers. The integration of scientific literacy is discussed first, followed by a discussion of each of the 5Es (Bybee et al. 2006). A timeframe for distributing the lesson phases is given; however, the activities included in this plan (see Supplementary Resources for specific lesson materials), should be modified to meet the needs and interest of students, and to align with allotted instructional time and objectives. 

Large language models have recently been able to perform well in a wide variety of circumstances. In this work, we explore the possi- bility of large language models, specifically GPT-3, to write explanations for middle-school mathematics problems, with the goal of eventually us- ing this process to rapidly generate explanations for the mathematics problems of new curricula as they emerge, shortening the time to inte- grate new curricula into online learning platforms. To generate expla- nations, two approaches were taken. The first approach attempted to summarize the salient advice in tutoring chat logs between students and live tutors. The second approach attempted to generate explanations us- ing few-shot learning from explanations written by teachers for similar mathematics problems. After explanations were generated, a survey was used to compare their quality to that of explanations written by teachers. We test our methodology using the GPT-3 language model. Ultimately, the synthetic explanations were unable to outperform teacher written explanations. In the future more powerful large language models may be employed, and GPT-3 may still be effective as a tool to augment teach- ers’ process for writing explanations, rather than as a tool to replace them. The explanations, survey results, analysis code, and a dataset of tutoring chat logs are all available at https://osf.io/wh5n9/. 

This study presents the growth and characterization of an 8.1 μm-emitting, InGaAs/AlInAs/InP-based quantum cascade laser (QCL) formed on an InP-on-Si composite template by metalorganic chemical vapor deposition (MOCVD). First, for the composite-template formation, a GaAs buffer layer was grown by solid-source molecular-beam epitaxy on a commercial (001) GaP/Si substrate, thus forming a GaAs/GaP/Si template. Next, an InP metamorphic buffer layer (MBL) structure was grown atop the GaAs/GaP/Si template by MOCVD, followed by the MOCVD growth of the full QCL structure. The top-surface morphology of the GaAs/GaP/Si template before and after the InP MBL growth was assessed via atomic force microscopy, over a 100 μm2 area, and no antiphase domains were found. The average threading dislocation density (TDD) for the GaAs/GaP/Si template was found to be ∼1 × 109 cm−2, with a slightly lower defect density of ∼7.9 × 108 cm−2 after the InP MBL growth. The lasing performance of the QCL structure grown on Si was compared to that of its counterpart grown on InP native substrate and found to be quite similar. That is, the threshold-current density of the QCL on Si, for deep-etched ridge-guide devices with uncoated facets, is somewhat lower than that for its counterpart on native InP substrate, 1.50 vs 1.92 kA/cm2, while the maximum output power per facet is 1.64 vs 1.47 W. These results further demonstrate the resilience of QCLs to relatively high residual TDD values.