Reproduce “STUDY: Socially Aware Temporally Causal Decoder Recommender Systems” 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
Eltayeb Ahmed, Diana Mincu, Lauren Harrell, Katherine Heller, Subhrajit Roy
STUDY predicts students’ next audiobook interactions by processing classroom histories with a transformer whose attention mask allows each prediction to use only earlier-timestamped interactions, from the student or classmates. On held-out Learning Ally logs, it achieves 31.86% Hits@1, compared with 28.06% for an individual-history transformer. This is offline prediction of recorded interactions, not evidence of improved reading or learning.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.