Reproduce “Language Models for Page-Level Layout Decisions in E-commerce Search” 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
Varun Joshi, Eva C. Song, ChengXiang Zhai
The authors test whether page-context representations predict Add-to-Cart labels for Walmart search requests with a displayed recommendation stack. They encode query and page text with ModernBERT embeddings and train a classifier, comparing it with prompted judgments and prompt-feature classifiers. On randomly held-out requests, REC-Snn reaches ROC-AUC 0.8530 versus 0.7396 for PFC-Snn; this predicts proxy labels, not causal layout benefit.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.