{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QOLZPOJF2NOQ53VKYTWMBDQAF5","short_pith_number":"pith:QOLZPOJF","schema_version":"1.0","canonical_sha256":"839797b925d35d0eeeaac4ecc08e002f6e384842e095b051ce9e9bc463e8f7db","source":{"kind":"arxiv","id":"2506.23102","version":2},"attestation_state":"computed","paper":{"title":"Region-Aware Multimodal Large Language Model via SlowFast Tokenization and Pseudo-Mask Guidance for 3D CT Report Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"eess.IV","authors_text":"Dongyeong Kim, Hyungbin Park, Hyunseok Lim, JiHyun Kim, Jimin Sung, Jinyoung Seo, Namkug Kim, Sunggu Kyung, Wooyoung Jo, Yoojin Nam","submitted_at":"2025-06-29T06:08:55Z","abstract_excerpt":"Current CT report generation frameworks predominantly rely on global feature representations, often failing to capture region-specific details and potentially missing certain abnormalities. To overcome this limitation, we propose MedRegion-CT, a region-focused multimodal large language model framework featuring three key innovations. First, we revisit the SlowFast strategy to jointly model global and fine-grained information and adapt it to the medical domain via a Region-based SlowFast Tokenizer that extracts tokens guided by clinically meaningful regions. Second, generated pseudo-masks guide"},"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":"2506.23102","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"eess.IV","submitted_at":"2025-06-29T06:08:55Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"c03551530231f188bb5d48bbb0324d234f212c4a130b07927b2970e696f57ed5","abstract_canon_sha256":"17546b7c91bde99e8554c699a7af6385b875c7f8b3c01871d2ee28912fc06494"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T00:17:08.249955Z","signature_b64":"Lbe/2lOvhihMgHT2Uu2V1kIdu64ChOlQ11hUYRg4PZbLjhVIfE0G/nTeSzQBAi8NFwUMqRoYu70lSRaTpP+lAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"839797b925d35d0eeeaac4ecc08e002f6e384842e095b051ce9e9bc463e8f7db","last_reissued_at":"2026-07-01T00:17:08.249420Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T00:17:08.249420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Region-Aware Multimodal Large Language Model via SlowFast Tokenization and Pseudo-Mask Guidance for 3D CT Report Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"eess.IV","authors_text":"Dongyeong Kim, Hyungbin Park, Hyunseok Lim, JiHyun Kim, Jimin Sung, Jinyoung Seo, Namkug Kim, Sunggu Kyung, Wooyoung Jo, Yoojin Nam","submitted_at":"2025-06-29T06:08:55Z","abstract_excerpt":"Current CT report generation frameworks predominantly rely on global feature representations, often failing to capture region-specific details and potentially missing certain abnormalities. To overcome this limitation, we propose MedRegion-CT, a region-focused multimodal large language model framework featuring three key innovations. First, we revisit the SlowFast strategy to jointly model global and fine-grained information and adapt it to the medical domain via a Region-based SlowFast Tokenizer that extracts tokens guided by clinically meaningful regions. Second, generated pseudo-masks guide"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23102","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/2506.23102/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":"2506.23102","created_at":"2026-07-01T00:17:08.249488+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.23102v2","created_at":"2026-07-01T00:17:08.249488+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23102","created_at":"2026-07-01T00:17:08.249488+00:00"},{"alias_kind":"pith_short_12","alias_value":"QOLZPOJF2NOQ","created_at":"2026-07-01T00:17:08.249488+00:00"},{"alias_kind":"pith_short_16","alias_value":"QOLZPOJF2NOQ53VK","created_at":"2026-07-01T00:17:08.249488+00:00"},{"alias_kind":"pith_short_8","alias_value":"QOLZPOJF","created_at":"2026-07-01T00:17:08.249488+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.15997","citing_title":"Segmentation, Detection and Explanation: A Unified Framework for CT Appearance Reasoning","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10437","citing_title":"Enhancing Fine-Grained Spatial Grounding in 3D CT Report Generation via Discriminative Guidance","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5","json":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5.json","graph_json":"https://pith.science/api/pith-number/QOLZPOJF2NOQ53VKYTWMBDQAF5/graph.json","events_json":"https://pith.science/api/pith-number/QOLZPOJF2NOQ53VKYTWMBDQAF5/events.json","paper":"https://pith.science/paper/QOLZPOJF"},"agent_actions":{"view_html":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5","download_json":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5.json","view_paper":"https://pith.science/paper/QOLZPOJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.23102&json=true","fetch_graph":"https://pith.science/api/pith-number/QOLZPOJF2NOQ53VKYTWMBDQAF5/graph.json","fetch_events":"https://pith.science/api/pith-number/QOLZPOJF2NOQ53VKYTWMBDQAF5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5/action/storage_attestation","attest_author":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5/action/author_attestation","sign_citation":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5/action/citation_signature","submit_replication":"https://pith.science/pith/QOLZPOJF2NOQ53VKYTWMBDQAF5/action/replication_record"}},"created_at":"2026-07-01T00:17:08.249488+00:00","updated_at":"2026-07-01T00:17:08.249488+00:00"}