pith. sign in
Pith Number

pith:CHLSQ2OO

pith:2026:CHLSQ2OOEZL3LSZTTAXEIPZEHU
not attested not anchored not stored refs pending

Diagnosing Korean-Language LLM Political Bias via Census-Grounded Agent Simulation

Sungwoo Kang

Census-grounded agent simulations diagnose political biases in Korean LLMs and accurately predict real election outcomes.

arxiv:2605.18395 v1 · 2026-05-18 · cs.CY · cs.AI

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{CHLSQ2OOEZL3LSZTTAXEIPZEHU}

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

Dynamo-K accurately predicts 3/3 presidential winners including a 2.1%p MAE on the 2022 race with 0.73%p margin and correctly identifies the dominant party in a held-out local election.

C2weakest assumption

The assumption that census-grounded agent profiles and scenario prompts produce voter behavior distributions that are sufficiently representative of actual Korean electorate responses for the purpose of diagnosing model bias.

C3one line summary

Dynamo-K is a census-grounded agent simulation that diagnoses three failure modes in Korean LLMs' political predictions and shows calibration methods that improve accuracy on historical elections.

References

19 extracted · 0 resolved · 0 Pith anchors

[1] Dynamo: Large-scale political simulation with LLM agents 2026
[2] L. P. Argyle, E. C. Busby, N. Fulda, J. R. Gubler, C. Rytting, and D. Wingate. Out of one, many: Using language models to simulate human samples.Political Analysis, 31(3):337–351, 2023 2023
[3] M. D. Jenkins and H. J. Kim. The role of misogyny in the 2022 Korean presidential election: Under- standing the backlash against feminism in industrialized democracies.Journal of East Asian Studies, 2 2022 · doi:10.1017/jea.2024.11
[4] S. Feng, C. Y. Park, Y. Liu, and Y. Tsvetkov. From pretraining data to language models to downstream tasks: Tracking the trails of political biases leading to unfair NLP models. InProceedings of the 6 2023
[5] G. Deffuant, D. Neau, F. Amblard, and G. Weisbuch. Mixing beliefs among interacting agents.Advances in Complex Systems, 3(01n04):87–98, 2000 2000
Receipt and verification
First computed 2026-05-20T00:05:58.616595Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

11d72869ce2657b5cb33982e443f243d3fc4f806b538b5cb90d3f78b691e3949

Aliases

arxiv: 2605.18395 · arxiv_version: 2605.18395v1 · doi: 10.48550/arxiv.2605.18395 · pith_short_12: CHLSQ2OOEZL3 · pith_short_16: CHLSQ2OOEZL3LSZT · pith_short_8: CHLSQ2OO
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/CHLSQ2OOEZL3LSZTTAXEIPZEHU \
  | 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: 11d72869ce2657b5cb33982e443f243d3fc4f806b538b5cb90d3f78b691e3949
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "98f74b1801f553bf8e9dd8e336c256def41a25c7e608dbb0daa1f48ecc03c03e",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CY",
    "submitted_at": "2026-05-18T13:42:23Z",
    "title_canon_sha256": "63ae9df0108577604fcc866616e6d8d8bde6fc79acdb9cb68a1ae9126537009f"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.18395",
    "kind": "arxiv",
    "version": 1
  }
}