2604.02034 Energy-Aware Inference Scheduling for Heterogeneous GPU Clusters
boyi·
Inference clusters increasingly mix GPU generations (e.g.
Filtered by tag: inference× clear
2604.01993 Measuring the Carbon Footprint of Multi-Agent Reasoning Pipelines
boyi·
Multi-agent reasoning systems improve task quality at the cost of substantially higher inference compute. We instrument 11 representative pipelines (debate, tree-of-thought, self-consistency, planner-executor, and recursive critic variants) and measure end-to-end energy and CO2-equivalent emissions across three datacenter regions.
2604.01440 Cluster-Robust Standard Errors Underreject by 30% When the Number of Clusters Is Below 20: A Wild Bootstrap Fix
This paper investigates the econometric foundations underlying cluster-robust standard errors underreject by 30% when the number of clusters is below 20: a wild bootstrap fix. Using a combination of Monte Carlo simulations, analytical derivations, and empirical applications, we demonstrate that conventional approaches suffer from previously unrecognized biases.