{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NNIIV3DMJGYXU2NAIFN5PVT3EW","short_pith_number":"pith:NNIIV3DM","schema_version":"1.0","canonical_sha256":"6b508aec6c49b17a69a0415bd7d67b259d80fbf00e4d496e4532b7b7d2cd50b8","source":{"kind":"arxiv","id":"2506.06940","version":1},"attestation_state":"computed","paper":{"title":"Understanding Sharpness Dynamics in NN Training with a Minimalist Example: The Effects of Dataset Difficulty, Depth, Stochasticity, and More","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chulhee Yun, Geonhui Yoo, Minhak Song","submitted_at":"2025-06-07T22:35:13Z","abstract_excerpt":"When training deep neural networks with gradient descent, sharpness often increases -- a phenomenon known as progressive sharpening -- before saturating at the edge of stability. Although commonly observed in practice, the underlying mechanisms behind progressive sharpening remain poorly understood. In this work, we study this phenomenon using a minimalist model: a deep linear network with a single neuron per layer. We show that this simple model effectively captures the sharpness dynamics observed in recent empirical studies, offering a simple testbed to better understand neural network train"},"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":"2506.06940","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-07T22:35:13Z","cross_cats_sorted":[],"title_canon_sha256":"ef45e54d83fcce02807094be4c42935272910b35803afeef3c5461bcbc4470f8","abstract_canon_sha256":"763c2f417b6a5959871f7e1a6264f41e03c0f341c1cbc2a9eb8d44ea2c2d45cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:58.615683Z","signature_b64":"j2ccmRZ2yBrPSUxWxrfNGjNEhoYNh8oFjQSGu7eaYs3gQxNWmOP/ZWquDUEeCnAugp/KpFw9YR4b/V1mruwOBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b508aec6c49b17a69a0415bd7d67b259d80fbf00e4d496e4532b7b7d2cd50b8","last_reissued_at":"2026-07-05T11:17:58.615203Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:58.615203Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Sharpness Dynamics in NN Training with a Minimalist Example: The Effects of Dataset Difficulty, Depth, Stochasticity, and More","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chulhee Yun, Geonhui Yoo, Minhak Song","submitted_at":"2025-06-07T22:35:13Z","abstract_excerpt":"When training deep neural networks with gradient descent, sharpness often increases -- a phenomenon known as progressive sharpening -- before saturating at the edge of stability. Although commonly observed in practice, the underlying mechanisms behind progressive sharpening remain poorly understood. In this work, we study this phenomenon using a minimalist model: a deep linear network with a single neuron per layer. We show that this simple model effectively captures the sharpness dynamics observed in recent empirical studies, offering a simple testbed to better understand neural network train"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06940","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/2506.06940/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":"2506.06940","created_at":"2026-07-05T11:17:58.615274+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.06940v1","created_at":"2026-07-05T11:17:58.615274+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06940","created_at":"2026-07-05T11:17:58.615274+00:00"},{"alias_kind":"pith_short_12","alias_value":"NNIIV3DMJGYX","created_at":"2026-07-05T11:17:58.615274+00:00"},{"alias_kind":"pith_short_16","alias_value":"NNIIV3DMJGYXU2NA","created_at":"2026-07-05T11:17:58.615274+00:00"},{"alias_kind":"pith_short_8","alias_value":"NNIIV3DM","created_at":"2026-07-05T11:17:58.615274+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05219","citing_title":"Gradient Descent with Large Step Size Restores Symmetry in Deep Linear Networks with Multi-Pathway","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW","json":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW.json","graph_json":"https://pith.science/api/pith-number/NNIIV3DMJGYXU2NAIFN5PVT3EW/graph.json","events_json":"https://pith.science/api/pith-number/NNIIV3DMJGYXU2NAIFN5PVT3EW/events.json","paper":"https://pith.science/paper/NNIIV3DM"},"agent_actions":{"view_html":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW","download_json":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW.json","view_paper":"https://pith.science/paper/NNIIV3DM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.06940&json=true","fetch_graph":"https://pith.science/api/pith-number/NNIIV3DMJGYXU2NAIFN5PVT3EW/graph.json","fetch_events":"https://pith.science/api/pith-number/NNIIV3DMJGYXU2NAIFN5PVT3EW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW/action/storage_attestation","attest_author":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW/action/author_attestation","sign_citation":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW/action/citation_signature","submit_replication":"https://pith.science/pith/NNIIV3DMJGYXU2NAIFN5PVT3EW/action/replication_record"}},"created_at":"2026-07-05T11:17:58.615274+00:00","updated_at":"2026-07-05T11:17:58.615274+00:00"}