{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MHAHFD2PTMJZKQ46HQATWMLBFS","short_pith_number":"pith:MHAHFD2P","schema_version":"1.0","canonical_sha256":"61c0728f4f9b1395439e3c013b31612ca9226fba840f896f11b0e786d650d446","source":{"kind":"arxiv","id":"2503.19859","version":3},"attestation_state":"computed","paper":{"title":"An Overview of Low-Rank Structures in the Training and Adaptation of Large Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.SP","math.OC","stat.CO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Benjamin D. Haeffele, Can Yaras, Laura Balzano, Peng Wang, Qing Qu, Soo Min Kwon, Tianjiao Ding, Zhangyang Wang","submitted_at":"2025-03-25T17:26:09Z","abstract_excerpt":"The substantial computational demands of modern large-scale deep learning present significant challenges for efficient training and deployment. Recent research has revealed a widespread phenomenon wherein deep networks inherently learn low-rank structures in their weights and representations during training. This tutorial paper provides a comprehensive review of advances in identifying and exploiting these low-rank structures, bridging mathematical foundations with practical applications. We present two complementary theoretical perspectives on the emergence of low-rankness: viewing it through"},"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":"2503.19859","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-25T17:26:09Z","cross_cats_sorted":["eess.SP","math.OC","stat.CO","stat.ML"],"title_canon_sha256":"de645edeee434714c185db01bb04a35be040a23c7fbfd520de2fbb8a4a138bcb","abstract_canon_sha256":"9711d82abcfd1724ba3e5e1ce487ab16a7e24b98c50a69334c95810ba69c3e8b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-09T01:19:35.554430Z","signature_b64":"kjFLP/DXUKLxTs9wQ/QYK4mGbluCr6PksrErsoTx66DSvkNIj0MF3WOr+YoxuF+BRGU4AAYbtvW55ziEdwJnDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61c0728f4f9b1395439e3c013b31612ca9226fba840f896f11b0e786d650d446","last_reissued_at":"2026-07-09T01:19:35.553872Z","signature_status":"signed_v1","first_computed_at":"2026-07-09T01:19:35.553872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Overview of Low-Rank Structures in the Training and Adaptation of Large Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.SP","math.OC","stat.CO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Benjamin D. Haeffele, Can Yaras, Laura Balzano, Peng Wang, Qing Qu, Soo Min Kwon, Tianjiao Ding, Zhangyang Wang","submitted_at":"2025-03-25T17:26:09Z","abstract_excerpt":"The substantial computational demands of modern large-scale deep learning present significant challenges for efficient training and deployment. Recent research has revealed a widespread phenomenon wherein deep networks inherently learn low-rank structures in their weights and representations during training. This tutorial paper provides a comprehensive review of advances in identifying and exploiting these low-rank structures, bridging mathematical foundations with practical applications. We present two complementary theoretical perspectives on the emergence of low-rankness: viewing it through"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19859","kind":"arxiv","version":3},"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/2503.19859/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":"2503.19859","created_at":"2026-07-09T01:19:35.553940+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.19859v3","created_at":"2026-07-09T01:19:35.553940+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19859","created_at":"2026-07-09T01:19:35.553940+00:00"},{"alias_kind":"pith_short_12","alias_value":"MHAHFD2PTMJZ","created_at":"2026-07-09T01:19:35.553940+00:00"},{"alias_kind":"pith_short_16","alias_value":"MHAHFD2PTMJZKQ46","created_at":"2026-07-09T01:19:35.553940+00:00"},{"alias_kind":"pith_short_8","alias_value":"MHAHFD2P","created_at":"2026-07-09T01:19:35.553940+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2606.19993","citing_title":"Activation- and Influence-Aware Ranks (AIR): Function-Preserving SVD Compression for LLMs","ref_index":168,"is_internal_anchor":true},{"citing_arxiv_id":"2606.06814","citing_title":"The Effect of Training Task Diversity on In-Context Learning through the Lens of Low-Dimensional Subspaces","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18598","citing_title":"Pointwise Generalization in Deep Neural Networks","ref_index":164,"is_internal_anchor":true},{"citing_arxiv_id":"2603.00541","citing_title":"Spectral Condition for $\\mu$P under Width-Depth Scaling","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS","json":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS.json","graph_json":"https://pith.science/api/pith-number/MHAHFD2PTMJZKQ46HQATWMLBFS/graph.json","events_json":"https://pith.science/api/pith-number/MHAHFD2PTMJZKQ46HQATWMLBFS/events.json","paper":"https://pith.science/paper/MHAHFD2P"},"agent_actions":{"view_html":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS","download_json":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS.json","view_paper":"https://pith.science/paper/MHAHFD2P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.19859&json=true","fetch_graph":"https://pith.science/api/pith-number/MHAHFD2PTMJZKQ46HQATWMLBFS/graph.json","fetch_events":"https://pith.science/api/pith-number/MHAHFD2PTMJZKQ46HQATWMLBFS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS/action/storage_attestation","attest_author":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS/action/author_attestation","sign_citation":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS/action/citation_signature","submit_replication":"https://pith.science/pith/MHAHFD2PTMJZKQ46HQATWMLBFS/action/replication_record"}},"created_at":"2026-07-09T01:19:35.553940+00:00","updated_at":"2026-07-09T01:19:35.553940+00:00"}