{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WIWOEGJW2NWQBNOIZ3Y3SKJT4D","short_pith_number":"pith:WIWOEGJW","schema_version":"1.0","canonical_sha256":"b22ce21936d36d00b5c8cef1b92933e0d5ae65587abbbfbd863336d5e25d8ab7","source":{"kind":"arxiv","id":"2501.13198","version":3},"attestation_state":"computed","paper":{"title":"SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Deyu Meng, Hanspeter Pfister, Hongming Piao, Kede Ma, Long-Kai Huang, Renzhen Wang, Wanhua Li, Yichen Wu, Ying Wei","submitted_at":"2025-01-22T20:00:41Z","abstract_excerpt":"Continual Learning (CL) with foundation models has recently emerged as a promising paradigm to exploit abundant knowledge acquired during pre-training for tackling sequential tasks. However, existing prompt-based and Low-Rank Adaptation-based (LoRA-based) methods often require expanding a prompt/LoRA pool or retaining samples of previous tasks, which poses significant scalability challenges as the number of tasks grows. To address these limitations, we propose Scalable Decoupled LoRA (SD-LoRA) for class incremental learning, which continually separates the learning of the magnitude and directi"},"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":"2501.13198","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T20:00:41Z","cross_cats_sorted":[],"title_canon_sha256":"25573e8290b2772cfd00c3c96f59e9e4e9d21969d966682a07e5147ea6828db3","abstract_canon_sha256":"367614c48dfa3653d6be51d2b365f4f03ae6accd0f00ceab9b55768c48b8091d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:25:47.023092Z","signature_b64":"tV/RMfOaeKvGE59NzgeOmM5/rnqTCTz9FKlfBCmAocXghhSvMotXGDsibEyzzuPZT04p9aXwlIU5zpDqTT36Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b22ce21936d36d00b5c8cef1b92933e0d5ae65587abbbfbd863336d5e25d8ab7","last_reissued_at":"2026-07-05T10:25:47.022446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:25:47.022446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Deyu Meng, Hanspeter Pfister, Hongming Piao, Kede Ma, Long-Kai Huang, Renzhen Wang, Wanhua Li, Yichen Wu, Ying Wei","submitted_at":"2025-01-22T20:00:41Z","abstract_excerpt":"Continual Learning (CL) with foundation models has recently emerged as a promising paradigm to exploit abundant knowledge acquired during pre-training for tackling sequential tasks. However, existing prompt-based and Low-Rank Adaptation-based (LoRA-based) methods often require expanding a prompt/LoRA pool or retaining samples of previous tasks, which poses significant scalability challenges as the number of tasks grows. To address these limitations, we propose Scalable Decoupled LoRA (SD-LoRA) for class incremental learning, which continually separates the learning of the magnitude and directi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13198","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/2501.13198/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":"2501.13198","created_at":"2026-07-05T10:25:47.022522+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.13198v3","created_at":"2026-07-05T10:25:47.022522+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13198","created_at":"2026-07-05T10:25:47.022522+00:00"},{"alias_kind":"pith_short_12","alias_value":"WIWOEGJW2NWQ","created_at":"2026-07-05T10:25:47.022522+00:00"},{"alias_kind":"pith_short_16","alias_value":"WIWOEGJW2NWQBNOI","created_at":"2026-07-05T10:25:47.022522+00:00"},{"alias_kind":"pith_short_8","alias_value":"WIWOEGJW","created_at":"2026-07-05T10:25:47.022522+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01873","citing_title":"G2LoRA: Gradient Orthogonal Low-Rank Adaptation Framework for Graph Continual Learning on Text-Attributed Graphs","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D","json":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D.json","graph_json":"https://pith.science/api/pith-number/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/graph.json","events_json":"https://pith.science/api/pith-number/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/events.json","paper":"https://pith.science/paper/WIWOEGJW"},"agent_actions":{"view_html":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D","download_json":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D.json","view_paper":"https://pith.science/paper/WIWOEGJW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.13198&json=true","fetch_graph":"https://pith.science/api/pith-number/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/graph.json","fetch_events":"https://pith.science/api/pith-number/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/action/storage_attestation","attest_author":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/action/author_attestation","sign_citation":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/action/citation_signature","submit_replication":"https://pith.science/pith/WIWOEGJW2NWQBNOIZ3Y3SKJT4D/action/replication_record"}},"created_at":"2026-07-05T10:25:47.022522+00:00","updated_at":"2026-07-05T10:25:47.022522+00:00"}