{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4ENDEG2KERN2MUQATGGRMSKGUL","short_pith_number":"pith:4ENDEG2K","schema_version":"1.0","canonical_sha256":"e11a321b4a245ba65200998d164946a2e844129225c0bf07167652447822786a","source":{"kind":"arxiv","id":"2505.24399","version":1},"attestation_state":"computed","paper":{"title":"LightSAM: Parameter-Agnostic Sharpness-Aware Minimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Enhong Chen, Hao Sun, Li Shen, Nan Yin, Xiaochun Cao, Yifei Cheng","submitted_at":"2025-05-30T09:28:38Z","abstract_excerpt":"Sharpness-Aware Minimization (SAM) optimizer enhances the generalization ability of the machine learning model by exploring the flat minima landscape through weight perturbations. Despite its empirical success, SAM introduces an additional hyper-parameter, the perturbation radius, which causes the sensitivity of SAM to it. Moreover, it has been proved that the perturbation radius and learning rate of SAM are constrained by problem-dependent parameters to guarantee convergence. These limitations indicate the requirement of parameter-tuning in practical applications. In this paper, we propose th"},"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":"2505.24399","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T09:28:38Z","cross_cats_sorted":[],"title_canon_sha256":"dfd93fac172b3e218b276a2436f8c19ff086fad287783e759b8ad08ca96e8275","abstract_canon_sha256":"34c126c7cd1511ab29b7daecb5f2cbd45df0a1a8e5912c81133a67c39d2bd3b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:48.378709Z","signature_b64":"LJwWjlcdOvjv3OableSKo+ABnkuLb/eWbsuYc7Daz5PXa7MYsoPD2qnXJDjLi+oKebBOcRQtjeHvdD/YIF3bBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e11a321b4a245ba65200998d164946a2e844129225c0bf07167652447822786a","last_reissued_at":"2026-07-05T11:12:48.378186Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:48.378186Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LightSAM: Parameter-Agnostic Sharpness-Aware Minimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Enhong Chen, Hao Sun, Li Shen, Nan Yin, Xiaochun Cao, Yifei Cheng","submitted_at":"2025-05-30T09:28:38Z","abstract_excerpt":"Sharpness-Aware Minimization (SAM) optimizer enhances the generalization ability of the machine learning model by exploring the flat minima landscape through weight perturbations. Despite its empirical success, SAM introduces an additional hyper-parameter, the perturbation radius, which causes the sensitivity of SAM to it. Moreover, it has been proved that the perturbation radius and learning rate of SAM are constrained by problem-dependent parameters to guarantee convergence. These limitations indicate the requirement of parameter-tuning in practical applications. In this paper, we propose th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24399","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/2505.24399/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":"2505.24399","created_at":"2026-07-05T11:12:48.378254+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24399v1","created_at":"2026-07-05T11:12:48.378254+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24399","created_at":"2026-07-05T11:12:48.378254+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ENDEG2KERN2","created_at":"2026-07-05T11:12:48.378254+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ENDEG2KERN2MUQA","created_at":"2026-07-05T11:12:48.378254+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ENDEG2K","created_at":"2026-07-05T11:12:48.378254+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01827","citing_title":"Adaptive Sharpness-Aware Minimization with a Polyak-type Step size: A Theory-Grounded Scheduler","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL","json":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL.json","graph_json":"https://pith.science/api/pith-number/4ENDEG2KERN2MUQATGGRMSKGUL/graph.json","events_json":"https://pith.science/api/pith-number/4ENDEG2KERN2MUQATGGRMSKGUL/events.json","paper":"https://pith.science/paper/4ENDEG2K"},"agent_actions":{"view_html":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL","download_json":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL.json","view_paper":"https://pith.science/paper/4ENDEG2K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24399&json=true","fetch_graph":"https://pith.science/api/pith-number/4ENDEG2KERN2MUQATGGRMSKGUL/graph.json","fetch_events":"https://pith.science/api/pith-number/4ENDEG2KERN2MUQATGGRMSKGUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL/action/storage_attestation","attest_author":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL/action/author_attestation","sign_citation":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL/action/citation_signature","submit_replication":"https://pith.science/pith/4ENDEG2KERN2MUQATGGRMSKGUL/action/replication_record"}},"created_at":"2026-07-05T11:12:48.378254+00:00","updated_at":"2026-07-05T11:12:48.378254+00:00"}