Paper accepted at AIED 2026

Congrats to Sam and Phillip on getting “Using Poly-Encoders for Computationally Efficient Automated Creativity Assessment” accepted as a short paper at #AIED2026! The work explores using Poly-Encoders paired with BERT-family models to assess student creativity — matching LLM-level accuracy at a fraction of the compute cost, running in milliseconds on consumer CPUs. A meaningful step toward making AI-powered creativity assessment practical in real classrooms. Well deserved in a record-breaking competitive year!