Reproduce “Deliberate Latent Reasoning for Sequential 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
Haobo Zhang, Qiannan Zhu, Yutao Zhu, Zhicheng Dou
DESRec trains sequential recommenders to preserve useful intermediate reasoning states while using a fixed inference depth. During training it selects the state most similar to the ground-truth next-item embedding and distills its item-ranking distribution into the final state. On Yelp with SASRec, NDCG@10 is 3.93 versus 3.68 for ReaRec-PRL, a reported 6.79% gain in offline ranking.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.