Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring

arXiv:2604.19984 · cs.CY, cs.AI, cs.CL · Submitted 2026-04-21 · Read on arXiv

cs.CY, cs.AI, cs.CL

Submitted: 2026-04-21

Updated: 2026-09-07

Comments: Second version, 45 pages, EMNLP 2026

Code: https://github.com/speedyapply/JobSpy

License: http://creativecommons.org/licenses/by/4.0/

The gist: Research has documented LLMs' name-based bias in hiring and salary recommendations.

Terminology

Abstract

Research has documented LLMs' name-based bias in hiring and salary recommendations. In this paper, we instead consider a setting where LLMs generate candidate summaries for downstream assessment. In a large-scale controlled study, we analyze nearly one million resume summaries produced by 4 models under systematic race-gender name perturbations, using synthetic resumes and real-world job postings. By decomposing each summary into resume-grounded factual content and evaluative framing, we find that factual content remains largely stable, while evaluative language exhibits subtle name-conditioned variation concentrated in the extremes of the distribution, especially in open-source models. Our hiring simulation demonstrates how evaluative summary transforms directional harm into symmetric instability that might evade conventional fairness audit, highlighting a potential pathway for LLM-to-LLM automation bias.

Sources

Related papers