{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZGI5DAQDIT4OJ6KP76IXZ35HEU","short_pith_number":"pith:ZGI5DAQD","schema_version":"1.0","canonical_sha256":"c991d1820344f8e4f94fff917cefa725371c9c854ca92e102f932eef0ebe9279","source":{"kind":"arxiv","id":"2607.23835","version":1},"attestation_state":"computed","paper":{"title":"Consistent Evidence, Robust Recognition: Faithful Attribution Regularization under Geometric Transformations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiawei Liang, Jiazi Hu, Qunli Zhang, Ruoyu Chen, Shangquan Sun, Shiming Liu, Wei Wang, Xianghao Jiao, Xiaochun Cao","submitted_at":"2026-07-26T20:41:38Z","abstract_excerpt":"Attribution methods are widely used to characterize the evidence underlying model predictions, yet their potential to improve model behavior remains underexplored. Attribution inconsistency under label-preserving geometric transformations may indicate transformation-sensitive evidence reliance, motivating attribution regularization. However, such supervision is valid only when attribution faithfully reflects the evidence driving predictions. Existing self-supervised methods typically align gradient-based maps such as Grad-CAM, whose limited faithfulness means that attribution consistency need "},"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":"2607.23835","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-26T20:41:38Z","cross_cats_sorted":[],"title_canon_sha256":"055ad3b1bd53dee4732832548dd6c5cfceb61b5ce949614d76a4ac69aa0568b6","abstract_canon_sha256":"e30b457442bf14017855804b35237e615271f46fd21811db83b94d02638a52f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T01:23:13.546806Z","signature_b64":"G03SHuQ8iyqNJS1A9iCWtbK8IjhB653I4fmC7a6YuZVNseVpay1Hzsp3/rGzQP7OmYLtFrMDHK/P/Jg6SdOyDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c991d1820344f8e4f94fff917cefa725371c9c854ca92e102f932eef0ebe9279","last_reissued_at":"2026-07-28T01:23:13.545974Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T01:23:13.545974Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Consistent Evidence, Robust Recognition: Faithful Attribution Regularization under Geometric Transformations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiawei Liang, Jiazi Hu, Qunli Zhang, Ruoyu Chen, Shangquan Sun, Shiming Liu, Wei Wang, Xianghao Jiao, Xiaochun Cao","submitted_at":"2026-07-26T20:41:38Z","abstract_excerpt":"Attribution methods are widely used to characterize the evidence underlying model predictions, yet their potential to improve model behavior remains underexplored. Attribution inconsistency under label-preserving geometric transformations may indicate transformation-sensitive evidence reliance, motivating attribution regularization. However, such supervision is valid only when attribution faithfully reflects the evidence driving predictions. Existing self-supervised methods typically align gradient-based maps such as Grad-CAM, whose limited faithfulness means that attribution consistency need "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23835","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/2607.23835/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":"2607.23835","created_at":"2026-07-28T01:23:13.546417+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.23835v1","created_at":"2026-07-28T01:23:13.546417+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23835","created_at":"2026-07-28T01:23:13.546417+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZGI5DAQDIT4O","created_at":"2026-07-28T01:23:13.546417+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZGI5DAQDIT4OJ6KP","created_at":"2026-07-28T01:23:13.546417+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZGI5DAQD","created_at":"2026-07-28T01:23:13.546417+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/ZGI5DAQDIT4OJ6KP76IXZ35HEU","json":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU.json","graph_json":"https://pith.science/api/pith-number/ZGI5DAQDIT4OJ6KP76IXZ35HEU/graph.json","events_json":"https://pith.science/api/pith-number/ZGI5DAQDIT4OJ6KP76IXZ35HEU/events.json","paper":"https://pith.science/paper/ZGI5DAQD"},"agent_actions":{"view_html":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU","download_json":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU.json","view_paper":"https://pith.science/paper/ZGI5DAQD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.23835&json=true","fetch_graph":"https://pith.science/api/pith-number/ZGI5DAQDIT4OJ6KP76IXZ35HEU/graph.json","fetch_events":"https://pith.science/api/pith-number/ZGI5DAQDIT4OJ6KP76IXZ35HEU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU/action/storage_attestation","attest_author":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU/action/author_attestation","sign_citation":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU/action/citation_signature","submit_replication":"https://pith.science/pith/ZGI5DAQDIT4OJ6KP76IXZ35HEU/action/replication_record"}},"created_at":"2026-07-28T01:23:13.546417+00:00","updated_at":"2026-07-28T01:23:13.546417+00:00"}