{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JSUXLPK6G73KKCBNNLOG7GJWUQ","short_pith_number":"pith:JSUXLPK6","schema_version":"1.0","canonical_sha256":"4ca975bd5e37f6a5082d6adc6f9936a43972de1dca87d339eea2105de04d589f","source":{"kind":"arxiv","id":"2508.13880","version":1},"attestation_state":"computed","paper":{"title":"In-hoc Concept Representations to Regularise Deep Learning in Medical Imaging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Floris Six Dijkstra, Hoel Kervadec, Kristoffer Wickstr{\\o}m, Regina Beets-Tan, Valentina Corbetta, Wilson Silva","submitted_at":"2025-08-19T14:43:30Z","abstract_excerpt":"Deep learning models in medical imaging often achieve strong in-distribution performance but struggle to generalise under distribution shifts, frequently relying on spurious correlations instead of clinically meaningful features. We introduce LCRReg, a novel regularisation approach that leverages Latent Concept Representations (LCRs) (e.g., Concept Activation Vectors (CAVs)) to guide models toward semantically grounded representations. LCRReg requires no concept labels in the main training set and instead uses a small auxiliary dataset to synthesise high-quality, disentangled concept examples."},"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":"2508.13880","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-19T14:43:30Z","cross_cats_sorted":[],"title_canon_sha256":"f4fd7a2f78abff2af6743a8ddee37049bab71b466893bedc6a65f42fbc94cf65","abstract_canon_sha256":"a44885f4acc726033fa9041cdb730b80b5f08ce48ccfa806e202ae05b9d2aacd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:09.064585Z","signature_b64":"qv9TvDVyqqkdJg+4SUXxh25ayyQZ0lLxk1N1I9EBijHxgEiYI5LBCpTexDLkAMfBg5GPmEk1NJvLg/hzujn3Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ca975bd5e37f6a5082d6adc6f9936a43972de1dca87d339eea2105de04d589f","last_reissued_at":"2026-07-05T11:56:09.064101Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:09.064101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In-hoc Concept Representations to Regularise Deep Learning in Medical Imaging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Floris Six Dijkstra, Hoel Kervadec, Kristoffer Wickstr{\\o}m, Regina Beets-Tan, Valentina Corbetta, Wilson Silva","submitted_at":"2025-08-19T14:43:30Z","abstract_excerpt":"Deep learning models in medical imaging often achieve strong in-distribution performance but struggle to generalise under distribution shifts, frequently relying on spurious correlations instead of clinically meaningful features. We introduce LCRReg, a novel regularisation approach that leverages Latent Concept Representations (LCRs) (e.g., Concept Activation Vectors (CAVs)) to guide models toward semantically grounded representations. LCRReg requires no concept labels in the main training set and instead uses a small auxiliary dataset to synthesise high-quality, disentangled concept examples."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.13880","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/2508.13880/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":"2508.13880","created_at":"2026-07-05T11:56:09.064156+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.13880v1","created_at":"2026-07-05T11:56:09.064156+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.13880","created_at":"2026-07-05T11:56:09.064156+00:00"},{"alias_kind":"pith_short_12","alias_value":"JSUXLPK6G73K","created_at":"2026-07-05T11:56:09.064156+00:00"},{"alias_kind":"pith_short_16","alias_value":"JSUXLPK6G73KKCBN","created_at":"2026-07-05T11:56:09.064156+00:00"},{"alias_kind":"pith_short_8","alias_value":"JSUXLPK6","created_at":"2026-07-05T11:56:09.064156+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/JSUXLPK6G73KKCBNNLOG7GJWUQ","json":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ.json","graph_json":"https://pith.science/api/pith-number/JSUXLPK6G73KKCBNNLOG7GJWUQ/graph.json","events_json":"https://pith.science/api/pith-number/JSUXLPK6G73KKCBNNLOG7GJWUQ/events.json","paper":"https://pith.science/paper/JSUXLPK6"},"agent_actions":{"view_html":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ","download_json":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ.json","view_paper":"https://pith.science/paper/JSUXLPK6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.13880&json=true","fetch_graph":"https://pith.science/api/pith-number/JSUXLPK6G73KKCBNNLOG7GJWUQ/graph.json","fetch_events":"https://pith.science/api/pith-number/JSUXLPK6G73KKCBNNLOG7GJWUQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ/action/storage_attestation","attest_author":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ/action/author_attestation","sign_citation":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ/action/citation_signature","submit_replication":"https://pith.science/pith/JSUXLPK6G73KKCBNNLOG7GJWUQ/action/replication_record"}},"created_at":"2026-07-05T11:56:09.064156+00:00","updated_at":"2026-07-05T11:56:09.064156+00:00"}