Pith Number

pith:MGGLWU5U

pith:2026:MGGLWU5UKLRCFV5TEVT3NKOAOF

not attested not anchored not stored refs pending

Generative Synthetic Data for Causal Inference: Pitfalls, Remedies, and Opportunities

Yichen Xu

Fully generative synthetic tabular data often preserves predictive performance while distorting average treatment effect estimates, because prediction loss only weakly constrains the treatment contrast.

arxiv:2604.23904 v3 · 2026-04-26 · stat.ME · cs.AI · 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{MGGLWU5UKLRCFV5TEVT3NKOAOF}

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 show that fully generative tabular synthesizers, including GAN- and LLM-based models, can preserve predictive utility while distorting average treatment effect (ATE) estimates. The failure is structural: ATE preservation requires both a realistic covariate law and an accurate treatment-effect contrast, whereas prediction loss penalizes treatment-effect error only through an overlap-weighted term.

C2weakest assumption

The assumption that separating covariate generation from treatment and outcome modeling in the hybrid framework can be done without introducing new biases or distortions while still producing realistic data for causal inference.

C3one line summary

Fully generative synthetic data preserves predictive utility but distorts ATE estimates due to a structural mismatch with prediction loss; a hybrid framework separating covariate generation from causal mechanisms improves fidelity.

Cited by

1 paper in Pith

When Does Synthetic Data Augmentation Improve Score-Based Imbalanced Classification?

Receipt and verification

First computed	2026-07-17T00:20:47.328198Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

618cbb53b452e222d7b32567b6a9c071766b1b1fc1e7d850ead24eefc45e366b

Aliases

arxiv: 2604.23904 · arxiv_version: 2604.23904v3 · doi: 10.48550/arxiv.2604.23904 · pith_short_12: MGGLWU5UKLRC · pith_short_16: MGGLWU5UKLRCFV5T · pith_short_8: MGGLWU5U

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/MGGLWU5UKLRCFV5TEVT3NKOAOF \
  | 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: 618cbb53b452e222d7b32567b6a9c071766b1b1fc1e7d850ead24eefc45e366b

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "57a86ccb766ce6ae2057783515ca3c22237070ed30ae285af0b87a6c117dee5f",
    "cross_cats_sorted": [
      "cs.AI",
      "stat.ML"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "stat.ME",
    "submitted_at": "2026-04-26T22:38:23Z",
    "title_canon_sha256": "fa94d9b896e0fd379a1e07a33d6a8614f0b383be1c52d47d800c61fce6fe3ffc"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.23904",
    "kind": "arxiv",
    "version": 3
  }
}