Timing beats volume: AI study shows how learners align with course cues
A learning analytics study of 10,452 online learners shows that temporal alignment with system-generated pedagogical events, measured by a new ...
A learning analytics study of 10,452 online learners shows that temporal alignment with system-generated pedagogical events, measured by a new ...
A new distillation framework that adjusts its temperature per language and sample compresses a 560-million-parameter multilingual model into a 66-million-parameter ...
A new causal hypergraph state space model builds a digital twin of each learner that removes confounding bias during representation ...
A new conceptual framework proposes that primary schools could run small language models on low-cost local servers, keeping all student ...
A new multimodal AI framework ranks classroom features by importance with 88.6 percent prediction accuracy, identifying linguistic coherence, student engagement, ...
Researchers in Shanghai have open-sourced a 1.5-billion-parameter language model trained from scratch for Chinese education, complete with a 258,000-entry instruction ...
A new multimodal Transformer framework called MFT-Net dynamically re-weights clicks, speech, gestures, and facial expressions to recognize 18 categories of ...
A study finds that adding hierarchical relationships among knowledge components can improve knowledge-tracing predictions and clarify students’ learning difficulties.
© 2025 Scienmag - Science Magazine
© 2025 Scienmag - Science Magazine