mortgage-underwriting

Tag

Cards List
#mortgage-underwriting

Fair outputs, Biased Internals: Causal Potency and Asymmetry of Latent Bias in LLMs for High-Stakes Decisions

arXiv cs.AI · 2026-05-18 Cached

This paper studies how instruction-tuned LLMs can exhibit fair outputs while retaining biased internal representations in high-stakes decisions like mortgage underwriting, showing that these hidden biases are causally potent, asymmetric, and exploitable through activation steering.

0 favorites 0 likes
← Back to home

Submit Feedback