{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:IKNXX2QCGPXWVDNTQYY3DXPWYE","short_pith_number":"pith:IKNXX2QC","schema_version":"1.0","canonical_sha256":"429b7bea0233ef6a8db38631b1ddf6c1335fa545df18be6f9a25279c55df0d37","source":{"kind":"arxiv","id":"2312.13555","version":2},"attestation_state":"computed","paper":{"title":"CR-SAM: Curvature Regularized Sharpness-Aware Minimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Donald C. Wunsch, Tao Wu, Tie Luo","submitted_at":"2023-12-21T03:46:29Z","abstract_excerpt":"The capacity to generalize to future unseen data stands as one of the utmost crucial attributes of deep neural networks. Sharpness-Aware Minimization (SAM) aims to enhance the generalizability by minimizing worst-case loss using one-step gradient ascent as an approximation. However, as training progresses, the non-linearity of the loss landscape increases, rendering one-step gradient ascent less effective. On the other hand, multi-step gradient ascent will incur higher training cost. In this paper, we introduce a normalized Hessian trace to accurately measure the curvature of loss landscape on"},"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":"2312.13555","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-21T03:46:29Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"18d4f61407d519a8b6e7193c4f3149c11d7494418d80a0b6f128946a9b44ad5a","abstract_canon_sha256":"486288571f077da4d4deeb925f5b87a68c56f6f0c82223ffdd317df5e604c039"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:27:28.168435Z","signature_b64":"uLUIMoa5pJoSivy0BqWw4MO+ecBiFvSO0Ju60SE7rXT4nhibbCp58na+BZxFKlBJO2xznKcA2XgBdr5i+yqmBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"429b7bea0233ef6a8db38631b1ddf6c1335fa545df18be6f9a25279c55df0d37","last_reissued_at":"2026-07-05T07:27:28.167973Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:27:28.167973Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CR-SAM: Curvature Regularized Sharpness-Aware Minimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Donald C. Wunsch, Tao Wu, Tie Luo","submitted_at":"2023-12-21T03:46:29Z","abstract_excerpt":"The capacity to generalize to future unseen data stands as one of the utmost crucial attributes of deep neural networks. Sharpness-Aware Minimization (SAM) aims to enhance the generalizability by minimizing worst-case loss using one-step gradient ascent as an approximation. However, as training progresses, the non-linearity of the loss landscape increases, rendering one-step gradient ascent less effective. On the other hand, multi-step gradient ascent will incur higher training cost. In this paper, we introduce a normalized Hessian trace to accurately measure the curvature of loss landscape on"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.13555","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/2312.13555/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":"2312.13555","created_at":"2026-07-05T07:27:28.168029+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.13555v2","created_at":"2026-07-05T07:27:28.168029+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.13555","created_at":"2026-07-05T07:27:28.168029+00:00"},{"alias_kind":"pith_short_12","alias_value":"IKNXX2QCGPXW","created_at":"2026-07-05T07:27:28.168029+00:00"},{"alias_kind":"pith_short_16","alias_value":"IKNXX2QCGPXWVDNT","created_at":"2026-07-05T07:27:28.168029+00:00"},{"alias_kind":"pith_short_8","alias_value":"IKNXX2QC","created_at":"2026-07-05T07:27:28.168029+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.02190","citing_title":"KANs need curvature: penalties for compositional smoothness","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE","json":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE.json","graph_json":"https://pith.science/api/pith-number/IKNXX2QCGPXWVDNTQYY3DXPWYE/graph.json","events_json":"https://pith.science/api/pith-number/IKNXX2QCGPXWVDNTQYY3DXPWYE/events.json","paper":"https://pith.science/paper/IKNXX2QC"},"agent_actions":{"view_html":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE","download_json":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE.json","view_paper":"https://pith.science/paper/IKNXX2QC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.13555&json=true","fetch_graph":"https://pith.science/api/pith-number/IKNXX2QCGPXWVDNTQYY3DXPWYE/graph.json","fetch_events":"https://pith.science/api/pith-number/IKNXX2QCGPXWVDNTQYY3DXPWYE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE/action/storage_attestation","attest_author":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE/action/author_attestation","sign_citation":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE/action/citation_signature","submit_replication":"https://pith.science/pith/IKNXX2QCGPXWVDNTQYY3DXPWYE/action/replication_record"}},"created_at":"2026-07-05T07:27:28.168029+00:00","updated_at":"2026-07-05T07:27:28.168029+00:00"}