Signals 4 · free daily AI digest

The Audit Decides the Verdict: Instrument Effects Rival Demographic Bias in LLM Decision Audits

Paper recorded by Signals 4 on 2026-09-08 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-08 on arXiv · recorded by Signals 4 on 2026-09-09

Category: cs.AI · 人工智能 · first seen 2026-09-09

Abstract

Whether a language model looks demographically biased can depend on how the audit asks its question. A charitable-aid benchmark reports that the same models favor minority applicants when rating requests one at a time and penalize some when ranking side by side. We test whether that reversal generalizes to hiring, lending, and medical triage: 40,726 requests to five models, applications differing

Read on arXiv →

#160 most recent of 300 cs.AI papers we have recorded · ↑ newer: Training-Free Task Vectors for LLM Behavioral Control · ↓ older: Diffusion TV: Experiencing Diffusion Models through Tangible, Embodied
Cite this page: The Audit Decides the Verdict: Instrument Effects Rival Demographic Bias in LLM Decision Audits: the #160 most recent of 300 cs.AI papers we have recorded (as of 2026-09-08). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/the-audit-decides-the-verdict-instrument-effects-rival-demographic-bias-in-llm-d.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.AI papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions