{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CQYZ7IAWXDOI5RO4SA46NJNFNW","short_pith_number":"pith:CQYZ7IAW","canonical_record":{"source":{"id":"2406.09044","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T12:30:02Z","cross_cats_sorted":[],"title_canon_sha256":"80a4d0b7b32a7c9f0b831c2f2f5653760baa2e55b898bb619ec402aa48b80b2f","abstract_canon_sha256":"0060735c31919c93f8fe3e6851a3a929f40f4976dcd097f2b5fe9da1afb0db13"},"schema_version":"1.0"},"canonical_sha256":"14319fa016b8dc8ec5dc9039e6a5a56dbdc81f21ae8bfbe28db1bca09e3621d8","source":{"kind":"arxiv","id":"2406.09044","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.09044","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"arxiv_version","alias_value":"2406.09044v3","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.09044","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"pith_short_12","alias_value":"CQYZ7IAWXDOI","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"pith_short_16","alias_value":"CQYZ7IAWXDOI5RO4","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"pith_short_8","alias_value":"CQYZ7IAW","created_at":"2026-07-05T10:22:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CQYZ7IAWXDOI5RO4SA46NJNFNW","target":"record","payload":{"canonical_record":{"source":{"id":"2406.09044","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T12:30:02Z","cross_cats_sorted":[],"title_canon_sha256":"80a4d0b7b32a7c9f0b831c2f2f5653760baa2e55b898bb619ec402aa48b80b2f","abstract_canon_sha256":"0060735c31919c93f8fe3e6851a3a929f40f4976dcd097f2b5fe9da1afb0db13"},"schema_version":"1.0"},"canonical_sha256":"14319fa016b8dc8ec5dc9039e6a5a56dbdc81f21ae8bfbe28db1bca09e3621d8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:22:00.879366Z","signature_b64":"s7Md4RyJB5cgerTqx7DmdgeX2Qyx+bC+ySbjhzT7tT7VB+tbbiFxqDAQbEF4NiMQYn5l3kFDeAfP6rUVCXrrCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"14319fa016b8dc8ec5dc9039e6a5a56dbdc81f21ae8bfbe28db1bca09e3621d8","last_reissued_at":"2026-07-05T10:22:00.878708Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:22:00.878708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.09044","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:22:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7RrZvR8kGso5in9kDa1n1CjqcgKyNwwFHI4MTgcTXIIlKA+uirw3qLOoAxL21iFCtQfULBuEY9x60A5kzJ8pDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T15:15:29.307829Z"},"content_sha256":"25910ddde3bde0c10ddd83bbfb0c77a56486291f0b0276cca590c078e6351415","schema_version":"1.0","event_id":"sha256:25910ddde3bde0c10ddd83bbfb0c77a56486291f0b0276cca590c078e6351415"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CQYZ7IAWXDOI5RO4SA46NJNFNW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MiLoRA: Harnessing Minor Singular Components for Parameter-Efficient LLM Finetuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Guanhua Chen, Hanqing Wang, Shuo Wang, Yixia Li, Yun Chen","submitted_at":"2024-06-13T12:30:02Z","abstract_excerpt":"Efficient finetuning of large language models (LLMs) aims to adapt the LLMs with reduced computational and memory cost. Previous LoRA-based approaches initialize the low-rank matrices with Gaussian distribution and zero values while keeping the original weight matrices frozen. However, the trainable model parameters optimized in an unguided subspace might interfere with the well-learned subspace of the pretrained weight matrices. In this paper, we propose MiLoRA, a simple yet effective LLM finetuning approach that only updates the minor singular components of the weight matrix while keeping th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.09044","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/2406.09044/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":"2026-07-05T10:22:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oiSAe55U2vs/TH37DzFASwPTBblx15w7uFtsGVbKnzk97/ww4fR5M+BxLMIca4050rjMCmxXjl7aXNOW1jg6AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T15:15:29.308775Z"},"content_sha256":"7e2e1b63c294ddd538d90e792ef56adfc289ccc27c69eab4ab80ad9dad5419b9","schema_version":"1.0","event_id":"sha256:7e2e1b63c294ddd538d90e792ef56adfc289ccc27c69eab4ab80ad9dad5419b9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW/bundle.json","state_url":"https://pith.science/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW/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":"2026-08-05T15:15:29Z","links":{"resolver":"https://pith.science/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW","bundle":"https://pith.science/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW/bundle.json","state":"https://pith.science/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CQYZ7IAWXDOI5RO4SA46NJNFNW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CQYZ7IAWXDOI5RO4SA46NJNFNW","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":"0060735c31919c93f8fe3e6851a3a929f40f4976dcd097f2b5fe9da1afb0db13","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T12:30:02Z","title_canon_sha256":"80a4d0b7b32a7c9f0b831c2f2f5653760baa2e55b898bb619ec402aa48b80b2f"},"schema_version":"1.0","source":{"id":"2406.09044","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.09044","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"arxiv_version","alias_value":"2406.09044v3","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.09044","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"pith_short_12","alias_value":"CQYZ7IAWXDOI","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"pith_short_16","alias_value":"CQYZ7IAWXDOI5RO4","created_at":"2026-07-05T10:22:00Z"},{"alias_kind":"pith_short_8","alias_value":"CQYZ7IAW","created_at":"2026-07-05T10:22:00Z"}],"graph_snapshots":[{"event_id":"sha256:7e2e1b63c294ddd538d90e792ef56adfc289ccc27c69eab4ab80ad9dad5419b9","target":"graph","created_at":"2026-07-05T10:22:00Z","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/2406.09044/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Efficient finetuning of large language models (LLMs) aims to adapt the LLMs with reduced computational and memory cost. Previous LoRA-based approaches initialize the low-rank matrices with Gaussian distribution and zero values while keeping the original weight matrices frozen. However, the trainable model parameters optimized in an unguided subspace might interfere with the well-learned subspace of the pretrained weight matrices. In this paper, we propose MiLoRA, a simple yet effective LLM finetuning approach that only updates the minor singular components of the weight matrix while keeping th","authors_text":"Guanhua Chen, Hanqing Wang, Shuo Wang, Yixia Li, Yun Chen","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T12:30:02Z","title":"MiLoRA: Harnessing Minor Singular Components for Parameter-Efficient LLM Finetuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.09044","kind":"arxiv","version":3},"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":"sha256:25910ddde3bde0c10ddd83bbfb0c77a56486291f0b0276cca590c078e6351415","target":"record","created_at":"2026-07-05T10:22:00Z","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":"0060735c31919c93f8fe3e6851a3a929f40f4976dcd097f2b5fe9da1afb0db13","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T12:30:02Z","title_canon_sha256":"80a4d0b7b32a7c9f0b831c2f2f5653760baa2e55b898bb619ec402aa48b80b2f"},"schema_version":"1.0","source":{"id":"2406.09044","kind":"arxiv","version":3}},"canonical_sha256":"14319fa016b8dc8ec5dc9039e6a5a56dbdc81f21ae8bfbe28db1bca09e3621d8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"14319fa016b8dc8ec5dc9039e6a5a56dbdc81f21ae8bfbe28db1bca09e3621d8","first_computed_at":"2026-07-05T10:22:00.878708Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:22:00.878708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"s7Md4RyJB5cgerTqx7DmdgeX2Qyx+bC+ySbjhzT7tT7VB+tbbiFxqDAQbEF4NiMQYn5l3kFDeAfP6rUVCXrrCg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:22:00.879366Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.09044","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:25910ddde3bde0c10ddd83bbfb0c77a56486291f0b0276cca590c078e6351415","sha256:7e2e1b63c294ddd538d90e792ef56adfc289ccc27c69eab4ab80ad9dad5419b9"],"state_sha256":"960bc40b3374fab9fdbd896ae1dedc84dbde3505cb40a906110e557514c68adc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"csh50x8+2hysqS4g2fAg+G/p/6CAhBUFkG5dv48vwgv3S3CVgS8+X0T52Enf6HMRIOx/iOf+qilHxywhVCQ0Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T15:15:29.314379Z","bundle_sha256":"c49830d61f1d79a6468340ce42cb473f114437f519ae3c138b2edb53a2e100c9"}}