{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:72C4NE4UH7IYM47TUZEEFGGIYZ","short_pith_number":"pith:72C4NE4U","schema_version":"1.0","canonical_sha256":"fe85c693943fd18673f3a6484298c8c675239299c2bc5eb1a34263bbbb21c2a2","source":{"kind":"arxiv","id":"2608.09801","version":1},"attestation_state":"computed","paper":{"title":"Modern Backbones Improve Multi-task DETR for Mammography Classification and Lesion Localization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Cao Truong Tran, Dac Phu Ho, Dinh Tan Nguyen, Lan T Ho-Pham, Le-Hoang Nguyen, Liem Pham, Minh-Toan Dinh, Nguyen Quoc Khanh Le, Quang-Hien Kha, Sai Ho Ling, Xuan-Huy Nguyen","submitted_at":"2026-08-10T16:22:07Z","abstract_excerpt":"Joint exam-level prediction and candidate-region localization may improve the usefulness of AI support in mammography. We study this setting using a multi-task DETR framework, where shared representations support both image-level malignancy prediction and lesion localization, and evaluate its performance on OPTIMAM and a biopsy-confirmed SGM1k cohort. Across both datasets, modern backbones consistently outperformed older ResNet-style features, with ConvNeXtV2 and DINOv3 giving the strongest overall results, whereas MambaVision was less competitive. On OPTIMAM, ConvNeXtV2 achieved the best over"},"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":"2608.09801","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-10T16:22:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"da3a66d3112895a6d23c9a1efce3c0d7258a7fd89ff3ea299f19187581afbe5d","abstract_canon_sha256":"482843c026c9895a65e234854c9ffe6a67e0ffb49e72dc23f24c4c7f2acdd05c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T02:25:07.704536Z","signature_b64":"BeBL5gmPaw3f2jbo3pB1SMfqwsuxAUckrAIB/2PWUKiVJ4S4yJAjLdEsYNQn7zrAEbmbKi2T9b7E2iKi41HDCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe85c693943fd18673f3a6484298c8c675239299c2bc5eb1a34263bbbb21c2a2","last_reissued_at":"2026-08-11T02:25:07.703062Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T02:25:07.703062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modern Backbones Improve Multi-task DETR for Mammography Classification and Lesion Localization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Cao Truong Tran, Dac Phu Ho, Dinh Tan Nguyen, Lan T Ho-Pham, Le-Hoang Nguyen, Liem Pham, Minh-Toan Dinh, Nguyen Quoc Khanh Le, Quang-Hien Kha, Sai Ho Ling, Xuan-Huy Nguyen","submitted_at":"2026-08-10T16:22:07Z","abstract_excerpt":"Joint exam-level prediction and candidate-region localization may improve the usefulness of AI support in mammography. We study this setting using a multi-task DETR framework, where shared representations support both image-level malignancy prediction and lesion localization, and evaluate its performance on OPTIMAM and a biopsy-confirmed SGM1k cohort. Across both datasets, modern backbones consistently outperformed older ResNet-style features, with ConvNeXtV2 and DINOv3 giving the strongest overall results, whereas MambaVision was less competitive. On OPTIMAM, ConvNeXtV2 achieved the best over"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09801","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/2608.09801/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":"2608.09801","created_at":"2026-08-11T02:25:07.703660+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.09801v1","created_at":"2026-08-11T02:25:07.703660+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09801","created_at":"2026-08-11T02:25:07.703660+00:00"},{"alias_kind":"pith_short_12","alias_value":"72C4NE4UH7IY","created_at":"2026-08-11T02:25:07.703660+00:00"},{"alias_kind":"pith_short_16","alias_value":"72C4NE4UH7IYM47T","created_at":"2026-08-11T02:25:07.703660+00:00"},{"alias_kind":"pith_short_8","alias_value":"72C4NE4U","created_at":"2026-08-11T02:25:07.703660+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/72C4NE4UH7IYM47TUZEEFGGIYZ","json":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ.json","graph_json":"https://pith.science/api/pith-number/72C4NE4UH7IYM47TUZEEFGGIYZ/graph.json","events_json":"https://pith.science/api/pith-number/72C4NE4UH7IYM47TUZEEFGGIYZ/events.json","paper":"https://pith.science/paper/72C4NE4U"},"agent_actions":{"view_html":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ","download_json":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ.json","view_paper":"https://pith.science/paper/72C4NE4U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.09801&json=true","fetch_graph":"https://pith.science/api/pith-number/72C4NE4UH7IYM47TUZEEFGGIYZ/graph.json","fetch_events":"https://pith.science/api/pith-number/72C4NE4UH7IYM47TUZEEFGGIYZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ/action/storage_attestation","attest_author":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ/action/author_attestation","sign_citation":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ/action/citation_signature","submit_replication":"https://pith.science/pith/72C4NE4UH7IYM47TUZEEFGGIYZ/action/replication_record"}},"created_at":"2026-08-11T02:25:07.703660+00:00","updated_at":"2026-08-11T02:25:07.703660+00:00"}