Open-source research

AMBER

An adaptive meta-balancing framework for integrating heterogeneous graph signals in knowledge tracing.

Year
2025
Status
active
Focus
Python · Graph Learning · Knowledge Tracing

What it does

AMBER tackles imbalance between heterogeneous graph views used to model student learning. A pretrained dual-graph teacher provides ensemble knowledge while a meta-distillation mechanism adjusts the teacher according to student-model feedback.

The official implementation includes the model, preprocessing utilities, and experiment configuration used for the CIKM 2025 study.

Evidence and scope

This page is a concise guide to the work. Use the primary paper for the experimental setup, quantitative results, limitations, and formal claims; use the repository for the implementation and current reproduction instructions.