Reproduce “Learning to Rank Explanations for Recommendation with Language Models” 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
Ben Kabongo, Vincent Guigue
LMStaR ranks review-derived explanations for a user and target item by encoding their interaction history and item description together, then comparing that representation with separately encoded candidate statements. NDCG measures whether relevant statements rank highly. It leads NDCG@1 across Beauty, Toys, and Yelp, including 20.773 versus CLER’s 12.521 on Toys, but not NDCG@5 on Yelp; user usefulness remains untested.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.