2 papers
cs.CY2025
When the Past Misleads: Rethinking Training Data Expansion Under Temporal Distribution Shifts
Chengyuan Yao, Yunxuan Tang, Christopher Brooks +2
Predictive models are typically trained on historical data to predict future outcomes. While it is commonly assumed that training on more historical data would improve model perfor…
cs.OS2024
Mercury: QoS-Aware Tiered Memory System
Jiaheng Lu, Yiwen Zhang, Hasan Al Maruf +4
Memory tiering has received wide adoption in recent years as an effective solution to address the increasing memory demands of memory-intensive workloads. However, existing tiered…