{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:FXYT4Y2RMM6NOPP33MQEDVJS6K","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"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}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.22274","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"arxiv_version","alias_value":"2604.22274v7","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.22274","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_12","alias_value":"FXYT4Y2RMM6N","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_16","alias_value":"FXYT4Y2RMM6NOPP3","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_8","alias_value":"FXYT4Y2R","created_at":"2026-05-27T01:04:58Z"}],"graph_snapshots":[{"event_id":"sha256:47156364f9c890bf33e2a4dbd54d03fb4e7bfa5a866e8230201389bc8e21d458","target":"graph","created_at":"2026-05-27T01:04:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Counterfactual verification grounds open-vocabulary scene graph relations in visual evidence by checking score drops after targeted removal"}],"snapshot_sha256":"6743c336725e2c73c9d8b0a39ed4ad34741883083489b164a395f8240cdc9d9c"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-21T11:34:05.495212Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-20T00:07:19.623413Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.22274/integrity.json","findings":[],"snapshot_sha256":"4bb5205b0c827a6c8466847462e010dc5b9a92b3130b7a5921dbf6a1dc9ed955","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible and fine-grained relation phrases beyond a fixed predicate vocabulary. While recent vision-language models greatly expand the semantic coverage of SGG, they also introduce a critical reliability issue: predicted relations may be driven by language priors or object co-occurrence rather than grounded visual evidence. In this paper, we propose an evidence-rounded open-vocabulary SGG framework based on counterfactual relation verification. Instead of directly accepting plausible relation proposals, our method","authors_text":"Chenyu Liu, Ruohan Zhang, Siyuan Chen, Suiyang Guang","cross_cats":[],"headline":"Counterfactual verification grounds open-vocabulary scene graph relations in visual evidence by checking score drops after targeted removal","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-04-24T06:34:45Z","title":"CAGE-SGG: Counterfactual Active Graph Evidence for Open-Vocabulary Scene Graph Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.22274","kind":"arxiv","version":7},"verdict":{"created_at":"2026-05-13T07:53:41.448895Z","id":"ed9dc599-6a94-42eb-a9ac-3161f197fff3","model_set":{"reader":"grok-4.3"},"one_line_summary":"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.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Counterfactual verification grounds open-vocabulary scene graph relations in visual evidence by checking score drops after targeted removal","strongest_claim":"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.","weakest_assumption":"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."}},"verdict_id":"ed9dc599-6a94-42eb-a9ac-3161f197fff3"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:323e01031602fb23a452a884f1e873c7dc7b9e2252cbdc3aa36e8b701b0346b5","target":"record","created_at":"2026-05-27T01:04:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"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}},"canonical_sha256":"2df13e6351633cd73dfbdb2041d532f280209c7dd810fd206629b0175f3f3667","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2df13e6351633cd73dfbdb2041d532f280209c7dd810fd206629b0175f3f3667","first_computed_at":"2026-05-27T01:04:58.231813Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:04:58.231813Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YkWZoY+d/73BQkjXGfPfomEw5HUqUnpa2FIrfzSrtosQTux/627+08AEdcTLz2jyYbpfi6Zr6crGTOOgwimlDQ==","signature_status":"signed_v1","signed_at":"2026-05-27T01:04:58.232513Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.22274","source_kind":"arxiv","source_version":7}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:323e01031602fb23a452a884f1e873c7dc7b9e2252cbdc3aa36e8b701b0346b5","sha256:47156364f9c890bf33e2a4dbd54d03fb4e7bfa5a866e8230201389bc8e21d458"],"state_sha256":"09e064ea1d8b6aae012a778868a5ab23f001d50ec7e2111425693e4e0c912dab"}