pith:FXYT4Y2R
CAGE-SGG: Counterfactual Active Graph Evidence for Open-Vocabulary Scene Graph Generation
Counterfactual verification grounds open-vocabulary scene graph relations in visual evidence by checking score drops after targeted removal
arxiv:2604.22274 v7 · 2026-04-24 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{FXYT4Y2RMM6NOPP33MQEDVJS6K}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Experiments on conventional, open-vocabulary, and panoptic SGG benchmarks show that our method consistently improves standard recall-based metrics, unseen predicate generalization, and counterfactual grounding quality. These results demonstrate that moving from relation generation to relation verification leads to more reliable, interpretable, and evidence-grounded scene graphs.
The assumption that decomposing predicate phrases into soft evidence bases such as support, contact, containment, depth and state, combined with a counterfactual verifier that tests whether the relation score decreases when necessary evidence is removed, accurately isolates true visual support without missing confounding factors or introducing new biases from the removal process.
CAGE-SGG improves open-vocabulary scene graph generation by verifying candidate relations through counterfactual removal of specific visual evidence rather than relying on language priors.
Receipt and verification
| First computed | 2026-05-27T01:04:58.231813Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
2df13e6351633cd73dfbdb2041d532f280209c7dd810fd206629b0175f3f3667
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/FXYT4Y2RMM6NOPP33MQEDVJS6K \
| 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: 2df13e6351633cd73dfbdb2041d532f280209c7dd810fd206629b0175f3f3667
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "3f0364497ffec162e23f2a0ec7f2ddd480840ea19431359b4f248165bac6c0a7",
"cross_cats_sorted": [],
"license": "http://creativecommons.org/licenses/by-sa/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-04-24T06:34:45Z",
"title_canon_sha256": "f1e43d22a8e135fbf81aa80eb6b70996dbe2e1f42f8b571b1237408246efccfb"
},
"schema_version": "1.0",
"source": {
"id": "2604.22274",
"kind": "arxiv",
"version": 7
}
}