Reproduce “PACERec: Progressive Adaptation of LLM Item Embeddings with Collaborative Evidence for Recommendation” 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
Junzhao Huang, Xi Wu, Zihan Wang, Zijing Wang, Ming Wang, Kang Liu, Xiaocui Yang, Shi Feng and 4 more
PACERec adapts item representations for next-item recommendation by learning a shared, semantics-conditioned mapping from items with sufficiently supported interaction histories toward representations of their preceding interaction contexts. It applies this adjustment to sparse and zero-interaction items without item-specific vectors. Across three datasets and nine backbone settings, it improves all four reported metrics; evaluations are offline and limited to tested settings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.