{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:S5F2WAIB52WBJZNGTY2MGUHE6H","short_pith_number":"pith:S5F2WAIB","schema_version":"1.0","canonical_sha256":"974bab0101eeac14e5a69e34c350e4f1ff63ad6bae7b571eebdca18c1cfa8400","source":{"kind":"arxiv","id":"2506.10355","version":2},"attestation_state":"computed","paper":{"title":"TreeLoRA: Efficient Continual Learning via Layer-Wise LoRAs Guided by a Hierarchical Gradient-Similarity Tree","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Peng Zhao, Yuan-Ze Xu, Yu-Yang Qian, Zhen-Yu Zhang, Zhi-Hua Zhou","submitted_at":"2025-06-12T05:25:35Z","abstract_excerpt":"Many real-world applications collect data in a streaming environment, where learning tasks are encountered sequentially. This necessitates continual learning (CL) to update models online, enabling adaptation to new tasks while preserving past knowledge to prevent catastrophic forgetting. Nowadays, with the flourish of large pre-trained models (LPMs), efficiency has become increasingly critical for CL, due to their substantial computational demands and growing parameter sizes. In this paper, we introduce TreeLoRA (K-D Tree of Low-Rank Adapters), a novel approach that constructs layer-wise adapt"},"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":"2506.10355","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-12T05:25:35Z","cross_cats_sorted":[],"title_canon_sha256":"027a693c36a86fd7aeccbf225eaefb63b11015af2849bcff5cf5e6f8871a240b","abstract_canon_sha256":"3e87df64968b03f01fde445cd89ccf83a4cfd16304e2db78e9302b428996038d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-29T01:14:02.790618Z","signature_b64":"4rv7V+M1J6kuj8Wwue9MMTouph4Aa7t+hG7yCVwIAs8ALNcOU9gchCLi0pnHfXsYnxaoWOkorBV4TgpvQQnODw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"974bab0101eeac14e5a69e34c350e4f1ff63ad6bae7b571eebdca18c1cfa8400","last_reissued_at":"2026-06-29T01:14:02.790112Z","signature_status":"signed_v1","first_computed_at":"2026-06-29T01:14:02.790112Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TreeLoRA: Efficient Continual Learning via Layer-Wise LoRAs Guided by a Hierarchical Gradient-Similarity Tree","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Peng Zhao, Yuan-Ze Xu, Yu-Yang Qian, Zhen-Yu Zhang, Zhi-Hua Zhou","submitted_at":"2025-06-12T05:25:35Z","abstract_excerpt":"Many real-world applications collect data in a streaming environment, where learning tasks are encountered sequentially. This necessitates continual learning (CL) to update models online, enabling adaptation to new tasks while preserving past knowledge to prevent catastrophic forgetting. Nowadays, with the flourish of large pre-trained models (LPMs), efficiency has become increasingly critical for CL, due to their substantial computational demands and growing parameter sizes. In this paper, we introduce TreeLoRA (K-D Tree of Low-Rank Adapters), a novel approach that constructs layer-wise adapt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10355","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/2506.10355/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":"2506.10355","created_at":"2026-06-29T01:14:02.790170+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.10355v2","created_at":"2026-06-29T01:14:02.790170+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10355","created_at":"2026-06-29T01:14:02.790170+00:00"},{"alias_kind":"pith_short_12","alias_value":"S5F2WAIB52WB","created_at":"2026-06-29T01:14:02.790170+00:00"},{"alias_kind":"pith_short_16","alias_value":"S5F2WAIB52WBJZNG","created_at":"2026-06-29T01:14:02.790170+00:00"},{"alias_kind":"pith_short_8","alias_value":"S5F2WAIB","created_at":"2026-06-29T01:14:02.790170+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2605.15053","citing_title":"TFGN: Task-Free, Replay-Free Continual Pre-Training Without Catastrophic Forgetting at LLM Scale","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05732","citing_title":"CRAFT: Forgetting-Aware Intervention-Based Adaptation for Continual Learning","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05732","citing_title":"CRAFT: Forgetting-Aware Intervention-Based Adaptation for Continual Learning","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07766","citing_title":"Head Similarity: Modeling Structured Whole-Head Appearance Beyond Face Recognition","ref_index":54,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H","json":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H.json","graph_json":"https://pith.science/api/pith-number/S5F2WAIB52WBJZNGTY2MGUHE6H/graph.json","events_json":"https://pith.science/api/pith-number/S5F2WAIB52WBJZNGTY2MGUHE6H/events.json","paper":"https://pith.science/paper/S5F2WAIB"},"agent_actions":{"view_html":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H","download_json":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H.json","view_paper":"https://pith.science/paper/S5F2WAIB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.10355&json=true","fetch_graph":"https://pith.science/api/pith-number/S5F2WAIB52WBJZNGTY2MGUHE6H/graph.json","fetch_events":"https://pith.science/api/pith-number/S5F2WAIB52WBJZNGTY2MGUHE6H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H/action/storage_attestation","attest_author":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H/action/author_attestation","sign_citation":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H/action/citation_signature","submit_replication":"https://pith.science/pith/S5F2WAIB52WBJZNGTY2MGUHE6H/action/replication_record"}},"created_at":"2026-06-29T01:14:02.790170+00:00","updated_at":"2026-06-29T01:14:02.790170+00:00"}