3 papers
cs.CL2026
Can LLMs Hire Fairly? Racial Bias in Resume Screening
Zhenyu Gao, Wenxi Jiang, Yutong Yan
We audit fourteen mainstream large language models (LLMs) for hiring discrimination using the paired-resume methodology of Kline, Rose, and Walters (2022). The sole 2023-vintage mo…
q-fin.GN2026
Debiasing LLMs by Fine-tuning
Zhenyu Gao, Wenxi Jiang, Yutong Yan
Prior research shows that large language models (LLMs) exhibit systematic extrapolation bias when forming predictions from both experimental and real-world data, and that prompt-ba…
cs.LG2026
Beyond Importance Sampling: Rejection-Gated Policy Optimization
Ziwu Sun, Zhen Gao, Jiyong Zhang +1
We propose a new perspective on policy optimization: rather than reweighting all samples by their importance ratios, an optimizer should select which samples are trustworthy enough…