{"ID":2869014,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.16400","arxiv_id":"2509.16400","title":"'Rich Dad, Poor Lad': How do Large Language Models Contextualize Socioeconomic Factors in College Admission ?","abstract":"Large Language Models (LLMs) are increasingly involved in high-stakes domains, yet how they reason about socially sensitive decisions remains underexplored. We present a large-scale audit of LLMs' treatment of socioeconomic status (SES) in college admissions decisions using a novel dual-process framework inspired by cognitive science. Leveraging a synthetic dataset of 30,000 applicant profiles grounded in real-world correlations, we prompt 4 open-source LLMs (Qwen 2, Mistral v0.3, Gemma 2, Llama 3.1) under 2 modes: a fast, decision-only setup (System 1) and a slower, explanation-based setup (System 2). Results from 5 million prompts reveal that LLMs consistently favor low-SES applicants -- even when controlling for academic performance -- and that System 2 amplifies this tendency by explicitly invoking SES as compensatory justification, highlighting both their potential and volatility as decision-makers. We then propose DPAF, a dual-process audit framework to probe LLMs' reasoning behaviors in sensitive applications.","short_abstract":"Large Language Models (LLMs) are increasingly involved in high-stakes domains, yet how they reason about socially sensitive decisions remains underexplored. We present a large-scale audit of LLMs' treatment of socioeconomic status (SES) in college admissions decisions using a novel dual-process framework inspired by co...","url_abs":"https://arxiv.org/abs/2509.16400","url_pdf":"https://arxiv.org/pdf/2509.16400v1","authors":"[\"Huy Nghiem\",\"Phuong-Anh Nguyen-Le\",\"John Prindle\",\"Rachel Rudinger\",\"Hal Daumé\"]","published":"2025-09-19T20:22:53Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.CY\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}