Reproduce “PAIR: Preference-Aligned Intent Retrieval for Premium People Recommendation at LinkedIn” 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
Zhongkai Shangguan, Yanbin Kang, Yihan Cao, Ruijie Jiang, Qiannan Yin, Chang Yue, Zijian Zhou, Girish Kathalagiri and 2 more
PAIR improves LinkedIn Premium Browsemaps retrieval by learning from profiles members click among those shown and combining member profiles with summaries of recent clicks. Precomputed embeddings let its click history update without online large language model re-encoding. On LinkedIn’s offline benchmark, it reaches MAR@100 0.450, 56% above production Holistic-171; a Premium product test reports relative CTR and retention gains, limited to that product and short-term window.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.