Search for: All records

Automated assessment of open responses in K–12 science education poses significant challenges due to the multimodal nature of student work, which often integrates textual explanations, drawings, and handwritten elements. Traditional evaluation methods that focus solely on textual analysis fail to capture the full breadth of student reasoning and are susceptible to biases such as handwriting neatness or answer length. In this paper, we propose a novel LLM-augmented multimodal evaluation framework that addresses these limitations through a comprehensive, bias-corrected grading system. Our approach leverages LLMs to generate causal knowledge graphs that encapsulate the essential conceptual relationships in student responses, comparing these graphs with those derived automatically from the rubrics and submissions. Experimental results demonstrate that our framework improves grading accuracy and consistency over deep supervised learning and few-shot LLM baselines. 

Short answer assessment is a vital component of science education, allowing evaluation of students' complex three-dimensional understanding. Large language models (LLMs) that possess human-like ability in linguistic tasks are increasingly popular in assisting human graders to reduce their workload. However, LLMs' limitations in domain knowledge restrict their understanding in task-specific requirements and hinder their ability to achieve satisfactory performance. Retrieval-augmented generation (RAG) emerges as a promising solution by enabling LLMs to access relevant domain-specific knowledge during assessment. In this work, we propose an adaptive RAG framework for automated grading that dynamically retrieves and incorporates domain-specific knowledge based on the question and student answer context. Our approach combines semantic search and curated educational sources to retrieve valuable reference materials. Experimental results in a science education dataset demonstrate that our system achieves an improvement in grading accuracy compared to baseline LLM approaches. The findings suggest that RAG-enhanced grading systems can serve as reliable support with efficient performance gains. 

Introduction:The unexpected surge of respiratory syncytial virus (RSV) cases following pandemic phase of COVID-19 has drawn much public attention. Drawing on the latest antiviral research, revisiting this heightened annual outbreak of respiratory disease could lead to new treatments. The ability of sulfated polysaccharides to compete for a variety of viruses binding to cell surface heparan sulfate, suggests several drugs that might have therapeutic potential for targeting RSV–glycosaminoglycan interactions. Methods:In the current study, the binding affinity and kinetics of two RSV glycoproteins (RSV-G protein and RSV-F protein) to heparin were investigated by surface plasmon resonance. Furthermore, solution competition studies using heparin oligosaccharides of different lengths indicated that the binding of RSV-G protein to heparin is size-dependent, whereas RSV-F protein did not show any chain length preference. Results and discussion:The two RSV glycoproteins have slightly different preferences for heparin sulfation patterns, but theN-sulfo group in heparin was most critical for the binding of heparin to both RSV-G protein and RSV-F protein. Finally, pentosan polysulfate and mucopolysaccharide polysulfate were evaluated for their inhibition of the RSV-G protein and RSV-F protein–heparin interaction, and both highly negative compounds showed strong inhibition. 

Abstract SARS-CoV-2 receptor binding domains (RBDs) interact with both the ACE2 receptor and heparan sulfate on the surface of host cells to enhance SARS-CoV-2 infection. We show that suramin, a polysulfated synthetic drug, binds to the ACE2 receptor and heparan sulfate binding sites on the RBDs of wild-type, Delta, and Omicron variants. Specifically, heparan sulfate and suramin had enhanced preferential binding for Omicron RBD, and suramin is most potent against the live SARS-CoV-2 Omicron variant (B.1.1.529) when compared to wild type and Delta (B.1.617.2) variants in vitro. These results suggest that inhibition of live virus infection occurs through dual SARS-CoV-2 targets of S-protein binding and previously reported RNA-dependent RNA polymerase inhibition and offers the possibility for this and other polysulfated molecules to be used as potential therapeutic and prophylactic options against COVID-19. 

Involving students in scientific modeling practice is one of the most effective approaches to achieving the next generation science education learning goals. Given the complexity and multirepresentational features of scientific models, scoring student-developed models is time- and cost-intensive, remaining one of the most challenging assessment practices for science education. More importantly, teachers who rely on timely feedback to plan and adjust instruction are reluctant to use modeling tasks because they could not provide timely feedback to learners. This study utilized machine learn- ing (ML), the most advanced artificial intelligence (AI), to develop an approach to automatically score student- drawn models and their written descriptions of those models. We developed six modeling assessment tasks for middle school students that integrate disciplinary core ideas and crosscutting concepts with the modeling practice. For each task, we asked students to draw a model and write a description of that model, which gave students with diverse backgrounds an opportunity to represent their understanding in multiple ways. We then collected student responses to the six tasks and had human experts score a subset of those responses. We used the human-scored student responses to develop ML algorithmic models (AMs) and to train the computer. Validation using new data suggests that the machine-assigned scores achieved robust agreements with human consent scores. Qualitative analysis of student-drawn models further revealed five characteristics that might impact machine scoring accuracy: Alternative expression, confusing label, inconsistent size, inconsistent position, and redundant information. We argue that these five characteristics should be considered when developing machine-scorable modeling tasks.