STA-ViT: A Spatiotemporal Self-Attention Vision Transformer for Learning Behavior Recognition and Intervention. IJCAI 2025, 49 (7). https://doi.org/10.31449/inf.v49i7.8653.