Pith Number

pith:AYL5TBQ6

pith:2026:AYL5TBQ6TF2CFIM3JTD6O7CPNW

not attested not anchored not stored refs pending

Causal Bias Detection in Generative Artificial Intelligence

Drago Plecko

Generative AI fairness can be quantified by decomposing bias along causal pathways and by measuring how the model's mechanisms replace real-world ones.

arxiv:2605.11365 v2 · 2026-05-12 · cs.AI · cs.LG · stat.ML

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{AYL5TBQ6TF2CFIM3JTD6O7CPNW}

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

We formalize the problem of causal fairness in generative AI and unify it with the standard ML setting under a common theoretical framework. We then derive new causal decomposition results that enable granular quantification of fairness impacts along both (a) different causal pathways and (b) the replacement of real-world mechanisms by the generative model's mechanisms.

C2weakest assumption

That generative models implicitly construct beliefs about all causal mechanisms and that identification conditions exist allowing the new decompositions to be estimated from data without strong additional assumptions on the generative process.

C3one line summary

A causal framework unifies fairness analysis across generative AI and standard ML by deriving decompositions that separate biases along causal pathways and differences between real-world and model mechanisms.

Receipt and verification

First computed	2026-05-20T00:05:47.019598Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

0617d9861e997422a19b4cc7e77c4f6d9a25b2f7b2df6ad946e6a6444080207c

Aliases

arxiv: 2605.11365 · arxiv_version: 2605.11365v2 · doi: 10.48550/arxiv.2605.11365 · pith_short_12: AYL5TBQ6TF2C · pith_short_16: AYL5TBQ6TF2CFIM3 · pith_short_8: AYL5TBQ6

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/AYL5TBQ6TF2CFIM3JTD6O7CPNW \
  | 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: 0617d9861e997422a19b4cc7e77c4f6d9a25b2f7b2df6ad946e6a6444080207c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "7f95a82f51fe9f40a47617803e6dbfe162ab895b00ceac37451d80538b07913d",
    "cross_cats_sorted": [
      "cs.LG",
      "stat.ML"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-12T00:36:53Z",
    "title_canon_sha256": "679916e7375f8dd4e88c5d1fb21ea6671d712fdff0230b674c61a8f031e2924c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.11365",
    "kind": "arxiv",
    "version": 2
  }
}