{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:O3EFWXWZ435PMPFAS4RHSXBUVX","short_pith_number":"pith:O3EFWXWZ","schema_version":"1.0","canonical_sha256":"76c85b5ed9e6faf63ca09722795c34adf2de8ea6f03f66304c37870946ce6503","source":{"kind":"arxiv","id":"2410.16029","version":1},"attestation_state":"computed","paper":{"title":"Natural GaLore: Accelerating GaLore for memory-efficient LLM Training and Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.LG","authors_text":"Arijit Das","submitted_at":"2024-10-21T14:05:06Z","abstract_excerpt":"Training LLMs presents significant memory challenges due to growing size of data, weights, and optimizer states. Techniques such as data and model parallelism, gradient checkpointing, and offloading strategies address this issue but are often infeasible due to hardware constraints. To mitigate memory usage, alternative methods like Parameter-Efficient-Fine-Tuning (PEFT) and GaLore approximate weights or optimizer states. PEFT methods, such as LoRA, have gained popularity for fine-tuning LLMs, though they require a full-rank warm start. In contrast, GaLore allows full-parameter learning while b"},"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":"2410.16029","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-21T14:05:06Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"e0802c7544a1774c3768e1f23289775855a33674880e04097a4e81f708e9587d","abstract_canon_sha256":"1f75a2e7f47744fecb4945c44f488817253e959612db31ec67a262639c3e3dc5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:23:28.274418Z","signature_b64":"oz2kDxDleTygf6PNtjkLmV82O8vYiJ9tfPYSK9lAqAllLJXGISvtzfiADGhD9uw60GQaOnRTPG69+yMyjHnHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"76c85b5ed9e6faf63ca09722795c34adf2de8ea6f03f66304c37870946ce6503","last_reissued_at":"2026-07-05T09:23:28.273923Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:23:28.273923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Natural GaLore: Accelerating GaLore for memory-efficient LLM Training and Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.LG","authors_text":"Arijit Das","submitted_at":"2024-10-21T14:05:06Z","abstract_excerpt":"Training LLMs presents significant memory challenges due to growing size of data, weights, and optimizer states. Techniques such as data and model parallelism, gradient checkpointing, and offloading strategies address this issue but are often infeasible due to hardware constraints. To mitigate memory usage, alternative methods like Parameter-Efficient-Fine-Tuning (PEFT) and GaLore approximate weights or optimizer states. PEFT methods, such as LoRA, have gained popularity for fine-tuning LLMs, though they require a full-rank warm start. In contrast, GaLore allows full-parameter learning while b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.16029","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/2410.16029/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":"2410.16029","created_at":"2026-07-05T09:23:28.273980+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.16029v1","created_at":"2026-07-05T09:23:28.273980+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.16029","created_at":"2026-07-05T09:23:28.273980+00:00"},{"alias_kind":"pith_short_12","alias_value":"O3EFWXWZ435P","created_at":"2026-07-05T09:23:28.273980+00:00"},{"alias_kind":"pith_short_16","alias_value":"O3EFWXWZ435PMPFA","created_at":"2026-07-05T09:23:28.273980+00:00"},{"alias_kind":"pith_short_8","alias_value":"O3EFWXWZ","created_at":"2026-07-05T09:23:28.273980+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.22922","citing_title":"Scalable Parameter and Memory Efficient Pretraining for LLM: Recent Algorithmic Advances and Benchmarking","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX","json":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX.json","graph_json":"https://pith.science/api/pith-number/O3EFWXWZ435PMPFAS4RHSXBUVX/graph.json","events_json":"https://pith.science/api/pith-number/O3EFWXWZ435PMPFAS4RHSXBUVX/events.json","paper":"https://pith.science/paper/O3EFWXWZ"},"agent_actions":{"view_html":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX","download_json":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX.json","view_paper":"https://pith.science/paper/O3EFWXWZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.16029&json=true","fetch_graph":"https://pith.science/api/pith-number/O3EFWXWZ435PMPFAS4RHSXBUVX/graph.json","fetch_events":"https://pith.science/api/pith-number/O3EFWXWZ435PMPFAS4RHSXBUVX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX/action/storage_attestation","attest_author":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX/action/author_attestation","sign_citation":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX/action/citation_signature","submit_replication":"https://pith.science/pith/O3EFWXWZ435PMPFAS4RHSXBUVX/action/replication_record"}},"created_at":"2026-07-05T09:23:28.273980+00:00","updated_at":"2026-07-05T09:23:28.273980+00:00"}