Pith Number

pith:PDPDJOLO

pith:2026:PDPDJOLOR4INUTOSA76UXJPU5J

not attested not anchored not stored refs pending

Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebook

Hyunjin Hwang, Jaehyeok Lee, Jing Yao, JinYeong Bak, Roy Ka-Wei Lee, Xiaoyuan Yi, Xing Xie

DOVE evaluates LLM cultural alignment by mapping texts to a learned value codebook and comparing distributions with unbalanced optimal transport.

arxiv:2604.06210 v3 · 2026-03-16 · cs.CL · cs.AI · cs.CY · cs.LG

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{PDPDJOLOR4INUTOSA76UXJPU5J}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

DOVE achieves superior predictive validity, attaining a 31.56% correlation with downstream tasks, while maintaining high reliability with as few as 500 samples per culture.

C2weakest assumption

The value codebook constructed from 10K documents via rate-distortion optimization accurately captures true cultural value orientations rather than surface-level semantic patterns, and that distributional differences measured by unbalanced optimal transport reflect genuine alignment rather than generation style artifacts.

C3one line summary

DOVE constructs a value codebook via rate-distortion variational optimization from 10K documents and measures LLM-human cultural alignment through unbalanced optimal transport, showing 31.56% correlation with downstream tasks and reliability at 500 samples per culture.

Formal links

2 machine-checked theorem links

Cited by

1 paper in Pith

When AI Speaks, Whose Values Does It Express? A Cross-Cultural Audit of Individualism-Collectivism Bias in Large Language Models

Receipt and verification

First computed	2026-06-02T02:04:52.823079Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

78de34b96e8f10da4dd207fd4ba5f4ea6b7cff50edcc8007098581a173a8474b

Aliases

arxiv: 2604.06210 · arxiv_version: 2604.06210v3 · doi: 10.48550/arxiv.2604.06210 · pith_short_12: PDPDJOLOR4IN · pith_short_16: PDPDJOLOR4INUTOS · pith_short_8: PDPDJOLO

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/PDPDJOLOR4INUTOSA76UXJPU5J \
  | 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: 78de34b96e8f10da4dd207fd4ba5f4ea6b7cff50edcc8007098581a173a8474b

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "3f1e1e7207143d9fdcdb09dc9d467025029087d94f282a0902c8dc736ad09da2",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CY",
      "cs.LG"
    ],
    "license": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-03-16T08:33:10Z",
    "title_canon_sha256": "34a68779826653e2793cc7f0f5d9eda708370080df6fb08cad56e2f9428ec0b3"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.06210",
    "kind": "arxiv",
    "version": 3
  }
}