{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XAQSR45QTJRVSYZ2PXEKBVBUEH","short_pith_number":"pith:XAQSR45Q","schema_version":"1.0","canonical_sha256":"b82128f3b09a6359633a7dc8a0d43421e54d15f005279c0488cb04a932abc0a0","source":{"kind":"arxiv","id":"2310.14152","version":1},"attestation_state":"computed","paper":{"title":"Orthogonal Subspace Learning for Language Model Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Han Xia, Qiming Ge, Qi Zhang, Rong Bao, Rui Zheng, Tao Gui, Tianze Chen, Xiao Wang, Xuanjing Huang","submitted_at":"2023-10-22T02:23:44Z","abstract_excerpt":"Benefiting from massive corpora and advanced hardware, large language models (LLMs) exhibit remarkable capabilities in language understanding and generation. However, their performance degrades in scenarios where multiple tasks are encountered sequentially, also known as catastrophic forgetting. In this paper, we propose orthogonal low-rank adaptation (O-LoRA), a simple and efficient approach for continual learning in language models, effectively mitigating catastrophic forgetting while learning new tasks. Specifically, O-LoRA learns tasks in different (low-rank) vector subspaces that are kept"},"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":"2310.14152","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-22T02:23:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"44cff7bbe9c5f109031adb9241facaa4f67658f63f3199c8080d8a06b90f8408","abstract_canon_sha256":"f8ff0bd0d84f627ab064669e984e1627cdc886865f220f2248a49ed6d97e8eaf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:40.379491Z","signature_b64":"2QRO7P3uVPslnTZ3XU74qPqU4m4f2KFf36M4dpjfSmK2h9ceIkNmUnNXPnVO6xgcyqQxmka7FpGkp+DpbEfSBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b82128f3b09a6359633a7dc8a0d43421e54d15f005279c0488cb04a932abc0a0","last_reissued_at":"2026-07-05T07:03:40.378996Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:40.378996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Orthogonal Subspace Learning for Language Model Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Han Xia, Qiming Ge, Qi Zhang, Rong Bao, Rui Zheng, Tao Gui, Tianze Chen, Xiao Wang, Xuanjing Huang","submitted_at":"2023-10-22T02:23:44Z","abstract_excerpt":"Benefiting from massive corpora and advanced hardware, large language models (LLMs) exhibit remarkable capabilities in language understanding and generation. However, their performance degrades in scenarios where multiple tasks are encountered sequentially, also known as catastrophic forgetting. In this paper, we propose orthogonal low-rank adaptation (O-LoRA), a simple and efficient approach for continual learning in language models, effectively mitigating catastrophic forgetting while learning new tasks. Specifically, O-LoRA learns tasks in different (low-rank) vector subspaces that are kept"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.14152","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/2310.14152/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":"2310.14152","created_at":"2026-07-05T07:03:40.379056+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.14152v1","created_at":"2026-07-05T07:03:40.379056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.14152","created_at":"2026-07-05T07:03:40.379056+00:00"},{"alias_kind":"pith_short_12","alias_value":"XAQSR45QTJRV","created_at":"2026-07-05T07:03:40.379056+00:00"},{"alias_kind":"pith_short_16","alias_value":"XAQSR45QTJRVSYZ2","created_at":"2026-07-05T07:03:40.379056+00:00"},{"alias_kind":"pith_short_8","alias_value":"XAQSR45Q","created_at":"2026-07-05T07:03:40.379056+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23487","citing_title":"CADRE: Stable, Parameter Efficient Adaptation of Medical Vision Language Models with Bounded Forgetting and Prior Drift","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02020","citing_title":"Hidden Forgetting in Continual Multimodal Learning: When Accuracy Survives but Grounding Fails","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29498","citing_title":"Mask the Target: A Plug-and-Play Regularizer Against LoRA Forgetting","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2509.12958","citing_title":"Forget What's Sensitive, Remember What Matters: Token-Level Differential Privacy in Memory Sculpting for Continual Learning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2406.11354","citing_title":"Preserving Knowledge in Large Language Model with Model-Agnostic Self-Decompression","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2506.21035","citing_title":"Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2508.04227","citing_title":"Continual Learning for VLMs: A Survey and Taxonomy Beyond Forgetting","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20247","citing_title":"CP-MoE: Consistency-Preserving Mixture-of-Experts for Continual Learning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15053","citing_title":"TFGN: Task-Free, Replay-Free Continual Pre-Training Without Catastrophic Forgetting at LLM Scale","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08949","citing_title":"Muon-OGD: Muon-based Spectral Orthogonal Gradient Projection for LLM Continual Learning","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":175,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10973","citing_title":"Rotation-Preserving Supervised Fine-Tuning","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08949","citing_title":"Muon-OGD: Muon-based Spectral Orthogonal Gradient Projection for LLM Continual Learning","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02829","citing_title":"Compress Then Adapt? No, Do It Together via Task-aware Union of Subspaces","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH","json":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH.json","graph_json":"https://pith.science/api/pith-number/XAQSR45QTJRVSYZ2PXEKBVBUEH/graph.json","events_json":"https://pith.science/api/pith-number/XAQSR45QTJRVSYZ2PXEKBVBUEH/events.json","paper":"https://pith.science/paper/XAQSR45Q"},"agent_actions":{"view_html":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH","download_json":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH.json","view_paper":"https://pith.science/paper/XAQSR45Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.14152&json=true","fetch_graph":"https://pith.science/api/pith-number/XAQSR45QTJRVSYZ2PXEKBVBUEH/graph.json","fetch_events":"https://pith.science/api/pith-number/XAQSR45QTJRVSYZ2PXEKBVBUEH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH/action/storage_attestation","attest_author":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH/action/author_attestation","sign_citation":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH/action/citation_signature","submit_replication":"https://pith.science/pith/XAQSR45QTJRVSYZ2PXEKBVBUEH/action/replication_record"}},"created_at":"2026-07-05T07:03:40.379056+00:00","updated_at":"2026-07-05T07:03:40.379056+00:00"}