{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:M33BKTXFJ55ITRXPUV723HCBU7","short_pith_number":"pith:M33BKTXF","schema_version":"1.0","canonical_sha256":"66f6154ee54f7a89c6efa57fad9c41a7fc28f830c89ed12e0356642c8e17737b","source":{"kind":"arxiv","id":"2505.16531","version":2},"attestation_state":"computed","paper":{"title":"HOFT: Householder Orthogonal Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Albert Sanchis, Alejandro Moreno Arcas, Alfons Juan, Jorge Civera","submitted_at":"2025-05-22T11:20:35Z","abstract_excerpt":"Adaptation of foundation models using low-rank methods is a widespread approach. Another way to adapt these models is to employ orthogonal fine-tuning methods, which are less time and memory efficient despite their good generalization properties. In this work, we propose Householder Orthogonal Fine-tuning (HOFT), a novel orthogonal fine-tuning method that aims to alleviate time and space complexity. Moreover, some theoretical properties of the orthogonal fine-tuning paradigm are explored. From this exploration, Scaled Householder Orthogonal Fine-tuning (SHOFT) is proposed. Both HOFT and SHOFT "},"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":"2505.16531","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-22T11:20:35Z","cross_cats_sorted":[],"title_canon_sha256":"e1e9660c90504301b9113b3debce15dbc0e8e358624856e8a9dcf3ba6243890b","abstract_canon_sha256":"a4fb90fc7d339b3c3e92c8bd2397bae48744dbd706b3957ddd1b632b56f8e263"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:08:03.203714Z","signature_b64":"Yg2AzZ6F8L3qHztACWxsOPl/lqa92DA+JP15jVkDzJt5kSXHepiCyurObyqNzpJSNHJl0eUezVzxK1RhEqs+Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"66f6154ee54f7a89c6efa57fad9c41a7fc28f830c89ed12e0356642c8e17737b","last_reissued_at":"2026-07-05T12:08:03.203154Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:08:03.203154Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HOFT: Householder Orthogonal Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Albert Sanchis, Alejandro Moreno Arcas, Alfons Juan, Jorge Civera","submitted_at":"2025-05-22T11:20:35Z","abstract_excerpt":"Adaptation of foundation models using low-rank methods is a widespread approach. Another way to adapt these models is to employ orthogonal fine-tuning methods, which are less time and memory efficient despite their good generalization properties. In this work, we propose Householder Orthogonal Fine-tuning (HOFT), a novel orthogonal fine-tuning method that aims to alleviate time and space complexity. Moreover, some theoretical properties of the orthogonal fine-tuning paradigm are explored. From this exploration, Scaled Householder Orthogonal Fine-tuning (SHOFT) is proposed. Both HOFT and SHOFT "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16531","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/2505.16531/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":"2505.16531","created_at":"2026-07-05T12:08:03.203225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.16531v2","created_at":"2026-07-05T12:08:03.203225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16531","created_at":"2026-07-05T12:08:03.203225+00:00"},{"alias_kind":"pith_short_12","alias_value":"M33BKTXFJ55I","created_at":"2026-07-05T12:08:03.203225+00:00"},{"alias_kind":"pith_short_16","alias_value":"M33BKTXFJ55ITRXP","created_at":"2026-07-05T12:08:03.203225+00:00"},{"alias_kind":"pith_short_8","alias_value":"M33BKTXF","created_at":"2026-07-05T12:08:03.203225+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.15916","citing_title":"LoCO: Low-rank Compositional Rotation Fine-tuning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2601.00417","citing_title":"Deep Delta Learning","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7","json":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7.json","graph_json":"https://pith.science/api/pith-number/M33BKTXFJ55ITRXPUV723HCBU7/graph.json","events_json":"https://pith.science/api/pith-number/M33BKTXFJ55ITRXPUV723HCBU7/events.json","paper":"https://pith.science/paper/M33BKTXF"},"agent_actions":{"view_html":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7","download_json":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7.json","view_paper":"https://pith.science/paper/M33BKTXF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.16531&json=true","fetch_graph":"https://pith.science/api/pith-number/M33BKTXFJ55ITRXPUV723HCBU7/graph.json","fetch_events":"https://pith.science/api/pith-number/M33BKTXFJ55ITRXPUV723HCBU7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7/action/storage_attestation","attest_author":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7/action/author_attestation","sign_citation":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7/action/citation_signature","submit_replication":"https://pith.science/pith/M33BKTXFJ55ITRXPUV723HCBU7/action/replication_record"}},"created_at":"2026-07-05T12:08:03.203225+00:00","updated_at":"2026-07-05T12:08:03.203225+00:00"}