Reproduce “SurvEval: Listwise Re-Ranking Evaluator via Survival-Based Exposure and Multi-Objective Value Modeling” 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
YaChen Yan, Liubo Li
SurvEval addresses re-ranking by estimating the probability users reach each list position and each item’s conditional engagement value. It models sequential exits with position-specific hazards, then combines reaching probabilities and values across the list. On KuaiRand, it achieves 0.8523 NDCG versus CAVE’s 0.8450 and leads all four reported dataset-metric comparisons, but component contributions are not isolated in these offline results.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.