Reliable relevance modeling and ranking are pivotal to legal information retrieval (IR), where graded relevance judgments are indispensable yet prohibitively expensive to acquire. Despite their importance, expert-validated graded relevance benchmarks remain scarce, especially for non-English statutory-law jurisdictions where citation practices differ fundamentally from common-law paradigms. We introduce KCC, a publicly available and reusable benchmark resource for legal IR in Korean civil law. KCC is constructed from 38,372 publicly accessible court decisions and comprises 2,942 query–candidate case pairs annotated with expert-validated four-level relevance labels. These labels jointly capture factual similarity and legal reasoning, explicitly addressing the absence of systematic citation links in statutory-law systems. To facilitate reproducibility, we provide a detailed description of the dataset construction pipeline, annotation protocol, relevance criteria, and inter-annotator agreement statistics validated by legal professionals. The annotation guidelines are designed to be adaptable to other statutory-law and non-English legal IR settings, enabling future extensions beyond the current dataset. We further present a set of reference benchmark experiments using representative retrieval approaches, including traditional IR models, neural rankers, and LLM-based prompting methods, to indicate recommended usage and evaluation practices with KCC. We believe that KCC provides a long-term resource for the research community to study relevance modeling, ranking behavior, and evaluation methodology in domain-specific legal IR scenarios beyond English and common-law traditions.

The increasing utilization of newly emerging networks (e.g., private-5G) across industries underscores the need for accurate traffic prediction to manage network resources effectively. However, rapidly emerging networks face challenges in accurate prediction due to limited training data at the early stage and fluctuation in traffic load at the maintenance stage. In response, we propose ISOML (Inter-Service Online Meta-Learning), a novel traffic prediction pipeline designed for newly emerging networks. ISOML utilizes meta-learning to address data scarcity and employs the EWC (Elastic Weight Consolidation) for online learning to learn dynamics of traffic patterns. Experimental validation in real-world datasets demonstrates the efficacy of ISOML in predicting traffic for emerging network environments.