pith:BU3UEWO5
Training-Free Cultural Alignment of Large Language Models via Persona Disagreement
Disagreement among World Values Survey personas steers black-box LLMs toward country-specific cultural preferences at inference time.
arxiv:2605.10843 v2 · 2026-05-11 · cs.CL · cs.AI · cs.CY
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{BU3UEWO5ENNQLSHBXLMT2OUX4Y}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Across 20 countries and 7 open-weight backbones (2B--70B), DISCA reduces cultural misalignment on MultiTP by 10--24% on the six backbones >=3.8B, and 2--7% on open-ended scenarios, without changing any weights.
That within-country sociodemographic disagreement, when instantiated via World-Values-Survey-grounded persona agents, constitutes the primary and sufficient steering signal for correcting cultural misalignment in black-box LLMs.
DISCA uses disagreement among WVS-grounded persona panels to apply loss-averse logit corrections that reduce cultural misalignment by 10-24% on MultiTP for models 3.8B and larger, without weight changes.
Formal links
Receipt and verification
| First computed | 2026-05-20T00:05:46.985676Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
0d374259dd235b05c8e1bad93d3a97e61cbad5293bb2c96740e2c8ef8810673a
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/BU3UEWO5ENNQLSHBXLMT2OUX4Y \
| 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: 0d374259dd235b05c8e1bad93d3a97e61cbad5293bb2c96740e2c8ef8810673a
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "1187e762c30f58fb65ac401bd1abd805e793a35db9cc374086fa3e407aa93a9c",
"cross_cats_sorted": [
"cs.AI",
"cs.CY"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CL",
"submitted_at": "2026-05-11T16:55:16Z",
"title_canon_sha256": "b29eba42aa854c84f53d19899d87438e619c92290f7e49cc2be0695e0fd07885"
},
"schema_version": "1.0",
"source": {
"id": "2605.10843",
"kind": "arxiv",
"version": 2
}
}