4 papers
RetryBench: Separating Retry Permission from Recovery Outcomes in GPT Tool Use
You Liu, Yue Liu, Quanchao Lu
False retry permission (FRP) occurs when an agent retries a timed-out tool call without support from the visible contract. RetryBench separates this error from realized recovery. S…
OSCAR: Order-aware Sensitivity Correction for AI Rankings
You Liu, Yue Liu, Quanchao Lu +2
LLM judges evaluate candidate language models through pairwise comparisons of their responses. Judge-aware Bradley--Terry models aggregate pairwise comparisons by jointly estimatin…
Safe Screening Rules for Group SLOPE
Runxue Bao, Quanchao Lu, Yanfu Zhang
Variable selection is a challenging problem in high-dimensional sparse learning, especially when group structures exist. Group SLOPE performs well for the adaptive selection of gro…
Safe Screening Rules for Group OWL Models
Runxue Bao, Quanchao Lu, Yanfu Zhang
Group Ordered Weighted -Norm (Group OWL) regularized models have emerged as a useful procedure for high-dimensional sparse multi-task learning with correlated features. Prox…