Reproduce “Neighborhood-Aware Unlearning for Generative 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
Pierre Lubitzsch, Yubao Tang, Maarten de Rijke, Sebastian Schelter
Neighborhood-Aware Unlearning (NAU) reduces a promoted item's influence in generative recommendation while explicitly controlling probability assigned to nearby items that share semantic identifier tokens. It combines forgetting and retain repair with neighborhood mass control, whose direction can preserve or suppress neighbors. In spam-removal settings, it approaches retraining on exposure and utility, but remains approximate and can trade local utility for stronger suppression.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.