{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OOAUB6RGGLCIJ6I42WLJYCX7IM","short_pith_number":"pith:OOAUB6RG","schema_version":"1.0","canonical_sha256":"738140fa2632c484f91cd5969c0aff433be15565f939a08243c2c454ca334f9c","source":{"kind":"arxiv","id":"2402.01376","version":2},"attestation_state":"computed","paper":{"title":"LoTR: Low Tensor Rank Weight Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aleksandr Mikhalev, Daniel Bershatsky, Daria Cherniuk, Ivan Oseledets, Talgat Daulbaev","submitted_at":"2024-02-02T13:00:38Z","abstract_excerpt":"In this paper we generalize and extend an idea of low-rank adaptation (LoRA) of large language models (LLMs) based on Transformer architecture. Widely used LoRA-like methods of fine-tuning LLMs are based on matrix factorization of gradient update. We introduce LoTR, a novel approach for parameter-efficient fine-tuning of LLMs which represents a gradient update to parameters in a form of tensor decomposition. Low-rank adapter for each layer is constructed as a product of three matrices, and tensor structure arises from sharing left and right multipliers of this product among layers. Simultaneou"},"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":"2402.01376","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-02T13:00:38Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"41fb7aeb3ba6a722987a188ca48fedd293a514abffad52d024cfbe85507462ad","abstract_canon_sha256":"2f1d699bf6cd1be8aaa39fb4e8da88b6d5a72dbab77b9817fcf667ee2d6ae7c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:41:04.088744Z","signature_b64":"AFnBpf7x09MdcO2nRcCRYfE3hf60bzm0YYj9gTzd+XeQjWLQx3jpJwI6i33VYwioxutHgcsCYxPlO0JUaxqgAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"738140fa2632c484f91cd5969c0aff433be15565f939a08243c2c454ca334f9c","last_reissued_at":"2026-07-05T07:41:04.088277Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:41:04.088277Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LoTR: Low Tensor Rank Weight Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aleksandr Mikhalev, Daniel Bershatsky, Daria Cherniuk, Ivan Oseledets, Talgat Daulbaev","submitted_at":"2024-02-02T13:00:38Z","abstract_excerpt":"In this paper we generalize and extend an idea of low-rank adaptation (LoRA) of large language models (LLMs) based on Transformer architecture. Widely used LoRA-like methods of fine-tuning LLMs are based on matrix factorization of gradient update. We introduce LoTR, a novel approach for parameter-efficient fine-tuning of LLMs which represents a gradient update to parameters in a form of tensor decomposition. Low-rank adapter for each layer is constructed as a product of three matrices, and tensor structure arises from sharing left and right multipliers of this product among layers. Simultaneou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01376","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/2402.01376/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":"2402.01376","created_at":"2026-07-05T07:41:04.088343+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.01376v2","created_at":"2026-07-05T07:41:04.088343+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01376","created_at":"2026-07-05T07:41:04.088343+00:00"},{"alias_kind":"pith_short_12","alias_value":"OOAUB6RGGLCI","created_at":"2026-07-05T07:41:04.088343+00:00"},{"alias_kind":"pith_short_16","alias_value":"OOAUB6RGGLCIJ6I4","created_at":"2026-07-05T07:41:04.088343+00:00"},{"alias_kind":"pith_short_8","alias_value":"OOAUB6RG","created_at":"2026-07-05T07:41:04.088343+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08565","citing_title":"EinSort: Sorting is All We Need for Tensorizing LLM","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2509.03234","citing_title":"TeRA: Vector-based Random Tensor Network for High-Rank Adaptation of Large Language Models","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21905","citing_title":"Low-Rank Adaptation Redux for Large Models","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM","json":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM.json","graph_json":"https://pith.science/api/pith-number/OOAUB6RGGLCIJ6I42WLJYCX7IM/graph.json","events_json":"https://pith.science/api/pith-number/OOAUB6RGGLCIJ6I42WLJYCX7IM/events.json","paper":"https://pith.science/paper/OOAUB6RG"},"agent_actions":{"view_html":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM","download_json":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM.json","view_paper":"https://pith.science/paper/OOAUB6RG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.01376&json=true","fetch_graph":"https://pith.science/api/pith-number/OOAUB6RGGLCIJ6I42WLJYCX7IM/graph.json","fetch_events":"https://pith.science/api/pith-number/OOAUB6RGGLCIJ6I42WLJYCX7IM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM/action/storage_attestation","attest_author":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM/action/author_attestation","sign_citation":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM/action/citation_signature","submit_replication":"https://pith.science/pith/OOAUB6RGGLCIJ6I42WLJYCX7IM/action/replication_record"}},"created_at":"2026-07-05T07:41:04.088343+00:00","updated_at":"2026-07-05T07:41:04.088343+00:00"}