{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T6L6RL3U3N3QBRXKPUZHK3DZWW","short_pith_number":"pith:T6L6RL3U","schema_version":"1.0","canonical_sha256":"9f97e8af74db7700c6ea7d32756c79b5840587c8b5721dad60280cf980906e29","source":{"kind":"arxiv","id":"2603.28387","version":2},"attestation_state":"computed","paper":{"title":"The Scaffold Effect: How Prompt Framing Drives Apparent Multimodal Gains in Clinical VLM Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Doan Nam Long Vu, Simone Balloccu","submitted_at":"2026-03-30T12:58:10Z","abstract_excerpt":"Trustworthy clinical AI requires that performance gains reflect genuine evidence integration rather than surface-level artifacts. We evaluate 12 open-weight vision-language models (VLMs) on binary classification across two clinical neuroimaging cohorts, \\textsc{FOR2107} (affective disorders) and \\textsc{OASIS-3} (cognitive decline). Both datasets come with structural MRI data that carries no reliable individual-level diagnostic signal. Under these conditions, smaller VLMs exhibit gains of up to 58\\% F1 upon introduction of neuroimaging context, with distilled models becoming competitive with c"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2603.28387","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-30T12:58:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"696a6aa444c6f5b578686bf278a53391e5a5eef6c48606530a90db84afa1ac96","abstract_canon_sha256":"d7032490b7c7c1a52869dde1b333569d9fbfad7303c9f3002c4b657cab1dd26c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:53.615025Z","signature_b64":"w8CRlxG91TR3RYMuVsLqX1p642IKK3FC97VvuW5csPU5NQkrhKiwSG7btTAEi5eyUw/FRaJ5J/8koj9MJyapBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f97e8af74db7700c6ea7d32756c79b5840587c8b5721dad60280cf980906e29","last_reissued_at":"2026-06-19T16:12:53.614602Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:53.614602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Scaffold Effect: How Prompt Framing Drives Apparent Multimodal Gains in Clinical VLM Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Doan Nam Long Vu, Simone Balloccu","submitted_at":"2026-03-30T12:58:10Z","abstract_excerpt":"Trustworthy clinical AI requires that performance gains reflect genuine evidence integration rather than surface-level artifacts. We evaluate 12 open-weight vision-language models (VLMs) on binary classification across two clinical neuroimaging cohorts, \\textsc{FOR2107} (affective disorders) and \\textsc{OASIS-3} (cognitive decline). Both datasets come with structural MRI data that carries no reliable individual-level diagnostic signal. Under these conditions, smaller VLMs exhibit gains of up to 58\\% F1 upon introduction of neuroimaging context, with distilled models becoming competitive with c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.28387","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.28387/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2603.28387","created_at":"2026-06-19T16:12:53.614662+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.28387v2","created_at":"2026-06-19T16:12:53.614662+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.28387","created_at":"2026-06-19T16:12:53.614662+00:00"},{"alias_kind":"pith_short_12","alias_value":"T6L6RL3U3N3Q","created_at":"2026-06-19T16:12:53.614662+00:00"},{"alias_kind":"pith_short_16","alias_value":"T6L6RL3U3N3QBRXK","created_at":"2026-06-19T16:12:53.614662+00:00"},{"alias_kind":"pith_short_8","alias_value":"T6L6RL3U","created_at":"2026-06-19T16:12:53.614662+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW","json":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW.json","graph_json":"https://pith.science/api/pith-number/T6L6RL3U3N3QBRXKPUZHK3DZWW/graph.json","events_json":"https://pith.science/api/pith-number/T6L6RL3U3N3QBRXKPUZHK3DZWW/events.json","paper":"https://pith.science/paper/T6L6RL3U"},"agent_actions":{"view_html":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW","download_json":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW.json","view_paper":"https://pith.science/paper/T6L6RL3U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.28387&json=true","fetch_graph":"https://pith.science/api/pith-number/T6L6RL3U3N3QBRXKPUZHK3DZWW/graph.json","fetch_events":"https://pith.science/api/pith-number/T6L6RL3U3N3QBRXKPUZHK3DZWW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW/action/storage_attestation","attest_author":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW/action/author_attestation","sign_citation":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW/action/citation_signature","submit_replication":"https://pith.science/pith/T6L6RL3U3N3QBRXKPUZHK3DZWW/action/replication_record"}},"created_at":"2026-06-19T16:12:53.614662+00:00","updated_at":"2026-06-19T16:12:53.614662+00:00"}