Preprint
Aug 2026
Matched Excess-Outranker Regularization for Candidate-Set Interference in Continual Knowledge Graph Embedding
This work introduces Matched Excess-Outranker Regularization (MEOR), a host-level objective that compares smooth answer-relative newcomer pressure with score-blind, structurally matched old references and establishes candidate admission as a distinct source of continual rank loss.
Hao Ren, Jun-Bin Gao, Jiao-Jiao Jiang
· 0 citations