{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FRB55JS6PYD2YTO3CAFSFPMSMO","short_pith_number":"pith:FRB55JS6","schema_version":"1.0","canonical_sha256":"2c43dea65e7e07ac4ddb100b22bd9263bc83821289dd06b9ef931b87a0c07eb0","source":{"kind":"arxiv","id":"2407.04400","version":1},"attestation_state":"computed","paper":{"title":"Hard-Attention Gates with Gradient Routing for Endoscopic Image Computing","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"eess.IV","authors_text":"Andrea Cherubini, Carlo Biffi, Giorgio Roffo, Pietro Salvagnini","submitted_at":"2024-07-05T10:20:24Z","abstract_excerpt":"To address overfitting and enhance model generalization in gastroenterological polyp size assessment, our study introduces Feature-Selection Gates (FSG) or Hard-Attention Gates (HAG) alongside Gradient Routing (GR) for dynamic feature selection. This technique aims to boost Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) by promoting sparse connectivity, thereby reducing overfitting and enhancing generalization. HAG achieves this through sparsification with learnable weights, serving as a regularization strategy. GR further refines this process by optimizing HAG parameters "},"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":"2407.04400","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"eess.IV","submitted_at":"2024-07-05T10:20:24Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"83091e1d8f1b55805ef3e4eee489ece1c366e54a74f279b3d520d4690a146bca","abstract_canon_sha256":"6a13575581d0511a5d8c2a44e880a2e1d9dd8add0e4bfee994d036f3cec70c3d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:40:33.370505Z","signature_b64":"bvx4YrokBynSRytuz9rnebrhsMnK7vwr/NpHKS/aIssGs5/uGlEsXoDLwv3VtDRL8H/tFpBLs3xlQ4pJKPpBBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c43dea65e7e07ac4ddb100b22bd9263bc83821289dd06b9ef931b87a0c07eb0","last_reissued_at":"2026-07-05T08:40:33.370111Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:40:33.370111Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hard-Attention Gates with Gradient Routing for Endoscopic Image Computing","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"eess.IV","authors_text":"Andrea Cherubini, Carlo Biffi, Giorgio Roffo, Pietro Salvagnini","submitted_at":"2024-07-05T10:20:24Z","abstract_excerpt":"To address overfitting and enhance model generalization in gastroenterological polyp size assessment, our study introduces Feature-Selection Gates (FSG) or Hard-Attention Gates (HAG) alongside Gradient Routing (GR) for dynamic feature selection. This technique aims to boost Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) by promoting sparse connectivity, thereby reducing overfitting and enhancing generalization. HAG achieves this through sparsification with learnable weights, serving as a regularization strategy. GR further refines this process by optimizing HAG parameters "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.04400","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/2407.04400/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":"2407.04400","created_at":"2026-07-05T08:40:33.370164+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.04400v1","created_at":"2026-07-05T08:40:33.370164+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.04400","created_at":"2026-07-05T08:40:33.370164+00:00"},{"alias_kind":"pith_short_12","alias_value":"FRB55JS6PYD2","created_at":"2026-07-05T08:40:33.370164+00:00"},{"alias_kind":"pith_short_16","alias_value":"FRB55JS6PYD2YTO3","created_at":"2026-07-05T08:40:33.370164+00:00"},{"alias_kind":"pith_short_8","alias_value":"FRB55JS6","created_at":"2026-07-05T08:40:33.370164+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.14560","citing_title":"The Origin of Self-Attention: Pairwise Affinity Matrices in Feature Selection and the Emergence of Self-Attention","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO","json":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO.json","graph_json":"https://pith.science/api/pith-number/FRB55JS6PYD2YTO3CAFSFPMSMO/graph.json","events_json":"https://pith.science/api/pith-number/FRB55JS6PYD2YTO3CAFSFPMSMO/events.json","paper":"https://pith.science/paper/FRB55JS6"},"agent_actions":{"view_html":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO","download_json":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO.json","view_paper":"https://pith.science/paper/FRB55JS6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.04400&json=true","fetch_graph":"https://pith.science/api/pith-number/FRB55JS6PYD2YTO3CAFSFPMSMO/graph.json","fetch_events":"https://pith.science/api/pith-number/FRB55JS6PYD2YTO3CAFSFPMSMO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO/action/storage_attestation","attest_author":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO/action/author_attestation","sign_citation":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO/action/citation_signature","submit_replication":"https://pith.science/pith/FRB55JS6PYD2YTO3CAFSFPMSMO/action/replication_record"}},"created_at":"2026-07-05T08:40:33.370164+00:00","updated_at":"2026-07-05T08:40:33.370164+00:00"}