Pith Number

pith:KWAHBRM7

pith:2026:KWAHBRM7BN5MUJRPYAJ5KTYFWL

not attested not anchored not stored refs resolved

Rethinking Generalization in Graph Neural Networks: A Structural Complexity Perspective

Jiye Liang, Liang Bai, Peiyao Wang, Richard Yi Da Xu, Xian Yang

More edges in a graph make GNN input representations overly accommodating to the model and induce overfitting.

arxiv:2605.13597 v1 · 2026-05-13 · 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{KWAHBRM7BN5MUJRPYAJ5KTYFWL}

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 theoretically prove that incorporating more edges into the prediction process transforms the input representations to be overly accommodating to the output model, thereby inducing overfitting... GNN generalization depends explicitly on structural complexity, alongside traditional parameter-dependent factors.

C2weakest assumption

That the number of effective edges (as defined in the structural complexity measure) is the dominant structural factor controlling generalization and that the Rademacher bound derived from it remains meaningful after the proposed regularization is applied.

C3one line summary

GNN generalization depends explicitly on graph structural complexity measured by effective edges, with a new regularization method shown to balance underfitting and overfitting.

References

61 extracted · 61 resolved · 2 Pith anchors

[1] Journal of Machine Learning Research , volume=

[2] Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence , volume=

[3] Journal of Machine Learning Research , volume=

[4] Proceedings of the Thirty-First ACM SIGKDD Conference on Knowledge Discovery and Data Mining , pages=

[5] Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence , volume=

Receipt and verification

First computed	2026-05-18T02:44:22.993265Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

558070c59f0b7aca262fc013d54f05b2f5eab7a3dc946b64195e7e520eeeb6d6

Aliases

arxiv: 2605.13597 · arxiv_version: 2605.13597v1 · doi: 10.48550/arxiv.2605.13597 · pith_short_12: KWAHBRM7BN5M · pith_short_16: KWAHBRM7BN5MUJRP · pith_short_8: KWAHBRM7

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/KWAHBRM7BN5MUJRPYAJ5KTYFWL \
  | 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: 558070c59f0b7aca262fc013d54f05b2f5eab7a3dc946b64195e7e520eeeb6d6

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "329373dee28472628ba5f1452fb6c598137acf030ef3ef7708fb9127e2790799",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-13T14:32:46Z",
    "title_canon_sha256": "b1b19cc5f3c0b0434bccc9d834d75428b5ab2a9e7e018b93f2d10676b4754ea5"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13597",
    "kind": "arxiv",
    "version": 1
  }
}