Search for: All records

We study the problem of Open-Vocabulary Constructs (OVCs)—ones not known beforehand—in the context of converting natural language (NL) specifications into formal languages (e.g., temporal logic or code). Mod- els fare poorly on OVCs due to a lack of necessary knowledge a priori. In such situations, a domain expert can provide correct constructs at in- ference time based on their preferences or domain knowledge. Our goal is to effectively reuse this inference-time, expert-provided knowledge for future parses without retraining the model. We present dynamic knowledge- augmented parsing (DKAP), where in addition to the input sentence, the model receives (dynamically growing) expert knowledge as a key-value lexicon that associates NL phrases with correct OVC constructs. We pro- pose ROLEX, a retrieval-augmented parsing approach that uses this lexicon. A retriever and a generator are trained to find and use the key-value store to produce the correct parse. A key challenge lies in curating data for this retrieval-augmented parser. We utilize synthetic data generation and the data augmentation techniques on annotated (NL sentence, FL statement) pairs to train the augmented parser. To improve training effectiveness, we propose multiple strategies to teach models to focus on the relevant subset of retrieved knowledge. Finally, we introduce a new evaluation paradigm modeled after the DKAP problem and simulate the scenario across three formalization tasks (NL2LTL, NL2Code, and NL2CMD). Our evaluations show that DKAP is a difficult challenge, and ROLEX helps improve the performance of baseline models by using dynamic expert knowledge effectively. 

Early forecasting of student performance in a course is a critical component of building effective intervention systems. However, when the available student data is limited, accurate early forecasting is challenging. We present a language generation transfer learning approach that leverages the general knowledge of pre-trained language models to address this challenge. We hypothesize that early forecasting can be significantly improved by fine-tuning large language models (LLMs) via personalization and contextualization using data on students' distal factors (academic and socioeconomic) and proximal non-cognitive factors (e.g., motivation and engagement), respectively. Results obtained from extensive experimentation validate this hypothesis and thereby demonstrate the prowess of personalization and contextualization for tapping into the general knowledge of pre-trained LLMs for solving the downstream task of early forecasting. 

In this paper, we present a framework for quantifying the impact of interventions on the full trajectories of students' experiences. The interventions are given periodically based on student performance forecasting from an artificial intelligence (AI) model. We performed a small-scale randomized controlled trial for evaluating the impact of the AI-based intervention system on the undergraduate students of a science, technology, engineering, and mathematics (STEM) course. Intervention messaging content was based on machine learning forecasting models trained on data collected from the students in the same course over the preceding 3 years. Trial results show that the intervention produced a statistically significant increase in the proportion of students that achieved a passing grade. By applying the trajectory-analysis framework we find that the intervention impacts the stories of some types of students more than others, and use this to define new ways of identifying students who are most likely to benefit. Together these outcomes point to the potential and promise of just-in-time interventions for STEM learning and the need for larger fully-powered randomized controlled trials. 

We present results from a small-scale randomized controlled trial that evaluates the impact of just-in-time interventions on the academic outcomes of N = 65 undergraduate students in a STEM course. Intervention messaging content was based on machine learning forecasting models of data collected from 537 students in the same course over the preceding 3 years. Trial results show that the intervention produced a statistically significant increase in the proportion of students that achieved a passing grade. The outcomes point to the potential and promise of just-in-time interventions for STEM learning and the need for larger fully-powered randomized controlled trials. 

A novel castor oil/water/ethanol Pickering emulsion, stabilized by magnetic nanoparticles (NPs), was developed to allow on-demand demulsification by an external magnetic field for the extraction of ethanol from aqueous solution using the castor oil. The emulsion was stabilized by Fe3O4-coated cellulose nanocrystals (CNC@Fe3O4) and lignin-coated Fe3O4 NPs (lignin@Fe3O4). The stability of the emulsions was investigated at various castor oil to ethanol-water ratios (50/50 and 70/30), various NP concentrations, and ethanol concentrations in the aqueous phase. The magnetically controlled demulsification ability of the emulsions was investigated by using a permanent magnet. The results showed that the 70/30 emulsions were more stable than the 50/50 emulsions for all the ethanol concentrations. Moreover, increasing the NP concentration increased the emulsion stability and hence, 1 w/v% NPs concentration provided the more stable systems. However, all the emulsions were successfully broken by the permanent magnet. Yet, the presence of ethanol improves the ability of the external magnetic field to demulsify these dispersions. Furthermore, the used hybrid NPs were recovered and recycled for three cycles. The recycled NPs were characterized with X-ray diffraction (XRD) and vibrating sample magnetometry (VSM) indicating that they retained their saturation magnetization and crystalline structure, demonstrating their lack of degradation over multiple recycling cycles. This study facilitates the exploration of innovative two-phase Pickering emulsions comprising three distinct liquid components and their utilization in liquid-liquid extraction processes. 

Nanoparticle additives increase the thermal conductivity of conventional heat transfer fluids at low concentrations, which leads to improved heat transfer fluids and processes. This study investigates lignin-coated magnetic nanocomposites (lignin@Fe3O4) as a novel bio-based magnetic nanoparticle additive to enhance the thermal conductivity of aqueous-based fluids. Kraft lignin was used to encapsulate the Fe3O4 nanoparticles to prevent agglomeration and oxidation of the magnetic nanoparticles. Lignin@Fe3O4 nanoparticles were prepared using a pH-driven co-precipitation method with a 3:1 lignin to magnetite ratio and characterized by X-ray diffraction, FT-IR, thermogravimetric analysis, and transmission electron microscopy. The magnetic properties were characterized using a vibrating sample magnetometer. Once fully characterized, lignin@Fe3O4 nanoparticles were dispersed in aqueous 0.1% w/v agar–water solutions at five different concentrations, from 0.001% w/v to 0.005% w/v. Thermal conductivity measurements were performed using the transient line heat source method at various temperatures. A maximum enhancement of 10% in thermal conductivity was achieved after adding 0.005% w/v lignin@Fe3O4 to the agar-based aqueous suspension at 45 °C. At room temperature (25 °C), the thermal conductivity of lignin@Fe3O4 and uncoated Fe3O4 agar-based suspensions was characterized at varying magnetic fields from 0 to 0.04 T, which were generated using a permanent magnet. For this analysis, the thermal conductivity of lignin magnetic nanosuspensions initially increased, showing a 5% maximum peak increase after applying a 0.02 T magnetic field, followed by a decreasing thermal conductivity at higher magnetic fields up to 0.04 T. This result is attributed to induced magnetic nanoparticle aggregation under external applied magnetic fields. Overall, this work demonstrates that lignin-coated Fe3O4 nanosuspension at low concentrations slightly increases the thermal conductivity of agar aqueous-based solutions, using a simple permanent magnet at room temperature or by adjusting temperature without any externally applied magnetic field.