{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XWCVTRZB5MYFDYTLGU6VHBI56K","short_pith_number":"pith:XWCVTRZB","schema_version":"1.0","canonical_sha256":"bd8559c721eb3051e26b353d53851df28428c1aea18cdcb003a4eeab40ea4354","source":{"kind":"arxiv","id":"2607.05438","version":1},"attestation_state":"computed","paper":{"title":"Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAG","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Xue Li, Yiming Gai","submitted_at":"2026-07-03T15:28:16Z","abstract_excerpt":"Multimodal retrieval-augmented generation (RAG) grounds a generator in evidence drawn from heterogeneous modalities -- text, tables, and images. The dominant deployment choice is binary and made before the model has tried to answer: either run a cheap text(+table) pipeline, or pay for an expensive vision-language model (VLM) over every image. Recent adaptive systems improve on this by selecting the modality or fidelity pre-retrieval, from a question-conditioned predictor of which modality will be needed. We show that this is the wrong decision point. Through an oracle headroom analysis on Mult"},"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":"2607.05438","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-07-03T15:28:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"418ec36602a42bb02200c9709604f7fe6a9de88ea23a69b9b7da6b9c0a1b7400","abstract_canon_sha256":"bf0e273a0e67e6c8f7d682f69e71ebe9a6182a4b898e4d057a02a6fb08927fed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:10.004134Z","signature_b64":"2kFNg79m8bK0fwdtH9NHw0sgFuivjwrN3Teq+Aur8v4hbA0xJt7SEIqgvdQnEfnFRV94fhWN+vWanwway+dDCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd8559c721eb3051e26b353d53851df28428c1aea18cdcb003a4eeab40ea4354","last_reissued_at":"2026-07-08T01:18:10.003730Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:10.003730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAG","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Xue Li, Yiming Gai","submitted_at":"2026-07-03T15:28:16Z","abstract_excerpt":"Multimodal retrieval-augmented generation (RAG) grounds a generator in evidence drawn from heterogeneous modalities -- text, tables, and images. The dominant deployment choice is binary and made before the model has tried to answer: either run a cheap text(+table) pipeline, or pay for an expensive vision-language model (VLM) over every image. Recent adaptive systems improve on this by selecting the modality or fidelity pre-retrieval, from a question-conditioned predictor of which modality will be needed. We show that this is the wrong decision point. Through an oracle headroom analysis on Mult"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05438","kind":"arxiv","version":1},"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/2607.05438/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":"2607.05438","created_at":"2026-07-08T01:18:10.003785+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.05438v1","created_at":"2026-07-08T01:18:10.003785+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05438","created_at":"2026-07-08T01:18:10.003785+00:00"},{"alias_kind":"pith_short_12","alias_value":"XWCVTRZB5MYF","created_at":"2026-07-08T01:18:10.003785+00:00"},{"alias_kind":"pith_short_16","alias_value":"XWCVTRZB5MYFDYTL","created_at":"2026-07-08T01:18:10.003785+00:00"},{"alias_kind":"pith_short_8","alias_value":"XWCVTRZB","created_at":"2026-07-08T01:18:10.003785+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/XWCVTRZB5MYFDYTLGU6VHBI56K","json":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K.json","graph_json":"https://pith.science/api/pith-number/XWCVTRZB5MYFDYTLGU6VHBI56K/graph.json","events_json":"https://pith.science/api/pith-number/XWCVTRZB5MYFDYTLGU6VHBI56K/events.json","paper":"https://pith.science/paper/XWCVTRZB"},"agent_actions":{"view_html":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K","download_json":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K.json","view_paper":"https://pith.science/paper/XWCVTRZB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.05438&json=true","fetch_graph":"https://pith.science/api/pith-number/XWCVTRZB5MYFDYTLGU6VHBI56K/graph.json","fetch_events":"https://pith.science/api/pith-number/XWCVTRZB5MYFDYTLGU6VHBI56K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K/action/storage_attestation","attest_author":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K/action/author_attestation","sign_citation":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K/action/citation_signature","submit_replication":"https://pith.science/pith/XWCVTRZB5MYFDYTLGU6VHBI56K/action/replication_record"}},"created_at":"2026-07-08T01:18:10.003785+00:00","updated_at":"2026-07-08T01:18:10.003785+00:00"}