Large Language Models (LLMs) have shown immense potential in education, automating tasks like quiz generation and content summarization. However, generating effective presentation slides introduces unique challenges due to the complexity of multimodal content creation and the need for precise, domain-specific information. Existing LLM-based solutions often fail to produce reliable and informative outputs, limiting their educational value. To address these limitations, we introduce SlideBot - a modular, multi-agent slide generation framework that integrates LLMs with retrieval, structured planning, and code generation. SlideBot is organized around three pillars: informativeness, ensuring deep and contextually grounded content; reliability, achieved by incorporating external sources through retrieval; and practicality, which enables customization and iterative feedback through instructor collaboration. It incorporates evidence-based instructional design principles from Cognitive Load Theory (CLT) and the Cognitive Theory of Multimedia Learning (CTML), using structured planning to manage intrinsic load and consistent visual macros to reduce extraneous load and enhance dual-channel learning. Within the system, specialized agents collaboratively retrieve information, summarize content, generate figures, and format slides using LaTeX, aligning outputs with instructor preferences through interactive refinement. Evaluations from domain experts and students in AI and biomedical education show that SlideBot consistently enhances conceptual accuracy, clarity, and instructional value. These findings demonstrate SlideBot’s potential to streamline slide preparation while ensuring accuracy, relevance, and adaptability in higher education.

Large language models (LLMs) have shown significant potential in scientific disciplines such as biomedicine, particularly in hypothesis generation, where they can analyze vast literature, identify patterns, and suggest research directions. However, a key challenge lies in evaluating the truthfulness of generated hypotheses, as verifying their accuracy often requires substantial time and resources. Additionally, the hallucination problem in LLMs can lead to the generation of hypotheses that appear plausible but are ultimately incorrect, undermining their reliability. To facilitate the systematic study of these challenges, we introduce TruthHypo, a benchmark for assessing the capabilities of LLMs in generating truthful scientific hypotheses, and KnowHD, a knowledge-based hallucination detector to evaluate how well hypotheses are grounded in existing knowledge. Our results show that LLMs struggle to generate truthful hypotheses. By analyzing hallucinations in reasoning steps, we demonstrate that the groundedness scores provided by KnowHD serve as an effective metric for filtering truthful hypotheses from the diverse outputs of LLMs. Human evaluations further validate the utility of KnowHD in identifying truthful hypotheses and accelerating scientific discovery. Our data and source code are available at https://github.com/Teddy-XiongGZ/TruthHypo.

Environmental stimuli evoke gene expression changes in neurons to achieve synaptic reconfiguration and memory formation. During the neuronal activation process, immediate early genes (IEGs) and late response genes (LRGs) have been defined according to their temporal changes in gene expression. Recent single cell studies indicated that different subtypes of neurons may have distinct sets of IEGs and LRGs in response to neuronal activation. Currently, it remains unexplored whether the gene expression patterns provided by single cell studies are sufficient to distinguish neuronal subtypes together with their activation states. In this study, we adopt Artificial Neural Network (ANN) models to predict neuron cell subtypes and activation states simultaneously based on a single cell RNA-Seq dataset generated from mouse visual cortex in response to light stimulus exposure. Four kinds of ANN architectures were designed to compare the prediction efficacies using two kinds of inputs: the full list of genes or the markers selected for neuron cell subtypes, IEGs, and LRGs. We observed that, with single-cell RNA-Seq data, neuron cell subtypes and activation states can be predicted accurately by ANN models. Moreover, compared to feeding the entire list of genes to fully connected ANNs, using selected marker genes and partially connected ANNs achieves comparable prediction accuracy with substantially improved computational efficiency. Altogether, our results demonstrated that ANNs are powerful in predicting neuron activation states and cell subtypes and may serve as a useful data mining approach to explore single cell RNA-Seq profiles.