Reproduce “CES-KD: Curriculum-based Expert Selection for Guided Knowledge Distillation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ibtihel Amara, Maryam Ziaeefard, Brett H. Meyer, Warren Gross, James J. Clark
CES-KD trains compact image classifiers by routing examples to teacher or assistant models according to difficulty. It ranks training images by a reference model’s loss, then assigns easier buckets to lower-capacity experts and harder buckets to larger ones; students also learn from labels. On one CINIC-10 setup, this assignment outscored reversed and random selection, without establishing broader superiority.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.