pith:PAEESUG7
When Fairness Metrics Disagree: Evaluating the Reliability of Demographic Fairness Assessment in Machine Learning
Different fairness metrics can lead to contradictory conclusions about bias in the same machine learning model.
arxiv:2604.15038 v2 · 2026-04-16 · cs.LG · cs.AI · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{PAEESUG733HZB4UVHFZ3RVSO32}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Our results demonstrate that fairness assessments can vary significantly depending on the choice of metrics, leading to contradictory conclusions regarding model bias.
That the selected fairness metrics, group partitions, and face-recognition setting are representative enough to generalize the disagreement finding to broader ML fairness evaluation.
Fairness metrics frequently disagree on bias levels in ML models, quantified by a new Fairness Disagreement Index that remains high across thresholds and configurations.
Cited by
Receipt and verification
| First computed | 2026-05-21T01:04:25.830809Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
78084950dfdecf90f2953973b8d64ede850742075d1722dbd1608a438e30758f
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/PAEESUG733HZB4UVHFZ3RVSO32 \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 78084950dfdecf90f2953973b8d64ede850742075d1722dbd1608a438e30758f
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "f4ca68405cb93eeaef69b4ac82fa8fe476fd5c0ef2c37d6ced48d4291c12093a",
"cross_cats_sorted": [
"cs.AI",
"cs.CV"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.LG",
"submitted_at": "2026-04-16T14:07:37Z",
"title_canon_sha256": "1f81ee62dac09bb08d7d7c8bb74028367860b15e91dc1e267bd4d3eee0abdcc6"
},
"schema_version": "1.0",
"source": {
"id": "2604.15038",
"kind": "arxiv",
"version": 2
}
}