pith:34NRRF6J
What Structural Inductive Bias Helps Transformers Reason Over Knowledge Graphs? A Study with Tabula RASA
Sparse adjacency masking alone supplies the main inductive bias that lets transformers perform multi-hop reasoning over knowledge graphs.
arxiv:2602.02834 v4 · 2026-02-02 · cs.LG · cs.AI
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{34NRRF6JQXR7AKCWNPE3MURFO7}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
sparse adjacency masking alone accounts for the dominant share of improvement over unmasked transformers (+72.5pp on 3-hop MetaQA, +45.5pp on WebQSP, +53.9pp on CWQ)
The four components can be removed independently without introducing implementation-specific interactions that confound the measured contributions of each.
Sparse adjacency masking alone explains the bulk of transformer gains on multi-hop KGQA tasks, showing that topological structure matters far more than relation-specific parameters.
Formal links
Receipt and verification
| First computed | 2026-06-04T01:08:41.239297Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
df1b1897c985e3f028566bc9b6522577e68b9aa52b6e0e5f2dad0c07e052081e
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/34NRRF6JQXR7AKCWNPE3MURFO7 \
| 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: df1b1897c985e3f028566bc9b6522577e68b9aa52b6e0e5f2dad0c07e052081e
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "bc1bdae5be0ae269d348759f2c9a2e5edead3ab5a3e9d8ee584711548d9440ea",
"cross_cats_sorted": [
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.LG",
"submitted_at": "2026-02-02T21:35:39Z",
"title_canon_sha256": "f539b0ff6befa3b748f7856c7d562b37480cacef14bbcd62bd35b265551c1281"
},
"schema_version": "1.0",
"source": {
"id": "2602.02834",
"kind": "arxiv",
"version": 4
}
}