Exact Risk Certificates for Pairwise-Judged Reranking
Overview & Synopsis
Pairwise relevance judgments constrain both the loss numerator and the ideal-ranking denominator of nDCG. We characterize the exact worst-case loss of replacing a ranking with a fixed proposal. Binary gains attain this loss and provide an inspectable witness. Threshold verification reduces to at most $K+1$ maximum-closure problems. Our exposure-flow ablation isolates denominator coupling: on shared judge trajectories, exact verification saves 2.34 comparisons per TREC-DL development query and 1.98 per confirmation query at a budget of 32. Savings transfer to BEIR and a second proposer. Predictive gates achieve higher deployment gains, while certificates expose the worst gains still permitted by accepted comparisons.
arxiv.org)
← / → for pages • + / - for zoom • 0 to fit width
Powered by Mozilla PDF.js v3.11
Document Information
| Document Title | Exact Risk Certificates for Pairwise-Judged Reranking |
|---|---|
| Author / Organization | Xinyu Wang, Jingrui Tian, Sicheng Lyu, Yu Gu, Ziyang Song, Xiao-Wen Chang, Lei Ding |
| Subject & Discipline | cs.LG |
| Publication Year | 2026 |
| License / Rights | arXiv Open Access / Creative Commons |
| Resource Verification | Listed PDF (Verification Pending) |
| Original Host Website |
arXiv
(arxiv.org)
|
| File Format | application/pdf (Portable Document Format) |
| Language | EN |
| Catalog ID | DOC-277740 |
| Original Source Link | Visit Source Page on arxiv.org » |