{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:5WKA67W4EJVWDQHOX7IZ76I7JT","short_pith_number":"pith:5WKA67W4","canonical_record":{"source":{"id":"2406.17296","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T05:45:12Z","cross_cats_sorted":[],"title_canon_sha256":"c07c4dacfcdd955edbb8b4dd80e77babef75749da43f9c3820ea5e90dd24d765","abstract_canon_sha256":"4362472a123d9f4e38d4ae98a277fc227707ee3b9add7296936b36234e27c478"},"schema_version":"1.0"},"canonical_sha256":"ed940f7edc226b61c0eebfd19ff91f4cf1fa6ecd0a029e733b380d292ac0fb2c","source":{"kind":"arxiv","id":"2406.17296","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.17296","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"arxiv_version","alias_value":"2406.17296v2","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17296","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"pith_short_12","alias_value":"5WKA67W4EJVW","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"pith_short_16","alias_value":"5WKA67W4EJVWDQHO","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"pith_short_8","alias_value":"5WKA67W4","created_at":"2026-07-05T09:49:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:5WKA67W4EJVWDQHOX7IZ76I7JT","target":"record","payload":{"canonical_record":{"source":{"id":"2406.17296","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T05:45:12Z","cross_cats_sorted":[],"title_canon_sha256":"c07c4dacfcdd955edbb8b4dd80e77babef75749da43f9c3820ea5e90dd24d765","abstract_canon_sha256":"4362472a123d9f4e38d4ae98a277fc227707ee3b9add7296936b36234e27c478"},"schema_version":"1.0"},"canonical_sha256":"ed940f7edc226b61c0eebfd19ff91f4cf1fa6ecd0a029e733b380d292ac0fb2c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:07.182631Z","signature_b64":"N53fH0kbgrBXJqfiMV937Ceaituhz3dXJzqBGRCMcck7niOKMn7BavfYSdfYWr6ajPtjZr6Ou8uQslKNaLszAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed940f7edc226b61c0eebfd19ff91f4cf1fa6ecd0a029e733b380d292ac0fb2c","last_reissued_at":"2026-07-05T09:49:07.182155Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:07.182155Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.17296","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:49:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H+dlFsk8TgB9qfmjFfScXGdlRnPdNuCVD5ukQV+ahoiPVwDyLs5afLNid1JZK76+NB29kTfiuiLwX3W/RQMmBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:38:07.773964Z"},"content_sha256":"79f47932d27cc2574d0aaeb9b705990a485ec5664e4fa4ecfa9992733a9a4128","schema_version":"1.0","event_id":"sha256:79f47932d27cc2574d0aaeb9b705990a485ec5664e4fa4ecfa9992733a9a4128"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:5WKA67W4EJVWDQHOX7IZ76I7JT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"BlockLLM: Memory-Efficient Adaptation of LLMs by Selecting and Optimizing the Right Coordinate Blocks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amrutha Varshini Ramesh, Issam H. Laradji, Mark Schmidt, Vignesh Ganapathiraman","submitted_at":"2024-06-25T05:45:12Z","abstract_excerpt":"Training large language models (LLMs) for pretraining or adapting to new tasks and domains has become increasingly critical as their applications expand. However, as the model and the data sizes grow, the training process presents significant memory challenges, often requiring a prohibitive amount of GPU memory that may not be readily available. Existing methods such as low-rank adaptation (LoRA) add trainable low-rank matrix factorizations, altering the training dynamics and limiting the model's parameter search to a low-rank subspace. GaLore, a more recent method, employs Gradient Low-Rank P"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17296","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/2406.17296/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:49:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uyLSewiUiDPZNzZf8pd9hq0JSDw0DF8VcLnlIzrnRSrriHYUfWs8mtfgLZqjhZrfNyQ11Tu7zttkatxlngmXBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:38:07.774345Z"},"content_sha256":"ded633f44e15cf9e833a84da9ed209e71fc50d83e6577e60108dca10357ce921","schema_version":"1.0","event_id":"sha256:ded633f44e15cf9e833a84da9ed209e71fc50d83e6577e60108dca10357ce921"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5WKA67W4EJVWDQHOX7IZ76I7JT/bundle.json","state_url":"https://pith.science/pith/5WKA67W4EJVWDQHOX7IZ76I7JT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5WKA67W4EJVWDQHOX7IZ76I7JT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-07T12:38:07Z","links":{"resolver":"https://pith.science/pith/5WKA67W4EJVWDQHOX7IZ76I7JT","bundle":"https://pith.science/pith/5WKA67W4EJVWDQHOX7IZ76I7JT/bundle.json","state":"https://pith.science/pith/5WKA67W4EJVWDQHOX7IZ76I7JT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5WKA67W4EJVWDQHOX7IZ76I7JT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:5WKA67W4EJVWDQHOX7IZ76I7JT","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"4362472a123d9f4e38d4ae98a277fc227707ee3b9add7296936b36234e27c478","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T05:45:12Z","title_canon_sha256":"c07c4dacfcdd955edbb8b4dd80e77babef75749da43f9c3820ea5e90dd24d765"},"schema_version":"1.0","source":{"id":"2406.17296","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.17296","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"arxiv_version","alias_value":"2406.17296v2","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17296","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"pith_short_12","alias_value":"5WKA67W4EJVW","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"pith_short_16","alias_value":"5WKA67W4EJVWDQHO","created_at":"2026-07-05T09:49:07Z"},{"alias_kind":"pith_short_8","alias_value":"5WKA67W4","created_at":"2026-07-05T09:49:07Z"}],"graph_snapshots":[{"event_id":"sha256:ded633f44e15cf9e833a84da9ed209e71fc50d83e6577e60108dca10357ce921","target":"graph","created_at":"2026-07-05T09:49:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2406.17296/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training large language models (LLMs) for pretraining or adapting to new tasks and domains has become increasingly critical as their applications expand. However, as the model and the data sizes grow, the training process presents significant memory challenges, often requiring a prohibitive amount of GPU memory that may not be readily available. Existing methods such as low-rank adaptation (LoRA) add trainable low-rank matrix factorizations, altering the training dynamics and limiting the model's parameter search to a low-rank subspace. GaLore, a more recent method, employs Gradient Low-Rank P","authors_text":"Amrutha Varshini Ramesh, Issam H. Laradji, Mark Schmidt, Vignesh Ganapathiraman","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T05:45:12Z","title":"BlockLLM: Memory-Efficient Adaptation of LLMs by Selecting and Optimizing the Right Coordinate Blocks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17296","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:79f47932d27cc2574d0aaeb9b705990a485ec5664e4fa4ecfa9992733a9a4128","target":"record","created_at":"2026-07-05T09:49:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"4362472a123d9f4e38d4ae98a277fc227707ee3b9add7296936b36234e27c478","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T05:45:12Z","title_canon_sha256":"c07c4dacfcdd955edbb8b4dd80e77babef75749da43f9c3820ea5e90dd24d765"},"schema_version":"1.0","source":{"id":"2406.17296","kind":"arxiv","version":2}},"canonical_sha256":"ed940f7edc226b61c0eebfd19ff91f4cf1fa6ecd0a029e733b380d292ac0fb2c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed940f7edc226b61c0eebfd19ff91f4cf1fa6ecd0a029e733b380d292ac0fb2c","first_computed_at":"2026-07-05T09:49:07.182155Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:49:07.182155Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"N53fH0kbgrBXJqfiMV937Ceaituhz3dXJzqBGRCMcck7niOKMn7BavfYSdfYWr6ajPtjZr6Ou8uQslKNaLszAw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:49:07.182631Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.17296","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79f47932d27cc2574d0aaeb9b705990a485ec5664e4fa4ecfa9992733a9a4128","sha256:ded633f44e15cf9e833a84da9ed209e71fc50d83e6577e60108dca10357ce921"],"state_sha256":"ae77d553e18d9fa1a6a244ca4666d6c36033f118ddf53b16993244a8d4c82b25"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OgOvQWFMOaQLpeKHirmtIshCAp+eT5t6VLaIUwGmDHy4ZttFCFHxkhDsQgq8UblWT5mRbwLbhBdfdnTf3sRdDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:38:07.776275Z","bundle_sha256":"090f85247dd03de6df48b52f62c1d161ef62011498fdd7dcac56766e4c279adb"}}