{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FZFLEPLAXJEK2RHAD76L7R6D2U","short_pith_number":"pith:FZFLEPLA","schema_version":"1.0","canonical_sha256":"2e4ab23d60ba48ad44e01ffcbfc7c3d5372ee8d9d7c04ab502caffa153976c29","source":{"kind":"arxiv","id":"2307.00526","version":2},"attestation_state":"computed","paper":{"title":"TensorGPT: Efficient Compression of Large Language Models based on Tensor-Train Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NA","cs.NE","math.NA"],"primary_cat":"cs.CL","authors_text":"Danilo P. Mandic, Mingxue Xu, Yao Lei Xu","submitted_at":"2023-07-02T09:33:09Z","abstract_excerpt":"High-dimensional token embeddings underpin Large Language Models (LLMs), as they can capture subtle semantic information and significantly enhance the modelling of complex language patterns. However, this high dimensionality also introduces considerable model parameters and prohibitively high model storage and memory requirements, which is particularly unaffordable for low-end devices. Targeting no extra training data and insufficient computation cases, we propose a training-free model compression approach based on the Tensor-Train Decomposition (TTD), whereby each pre-trained token embedding "},"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":"2307.00526","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-02T09:33:09Z","cross_cats_sorted":["cs.LG","cs.NA","cs.NE","math.NA"],"title_canon_sha256":"00950a38e1119152265253ef9cfb9df819fbab8fb36e93d750faadfccd73670a","abstract_canon_sha256":"0a5bcefab8d23bd221a315ae19fc34c6a763ad9eca4c5b64c1217739a3c2ee5a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:26.485526Z","signature_b64":"SRFAxdfZ0ZGAw4ARZnK5DxPxmG2jF9v5pi70hdrduUo7AQD2pM7V2hJqgIrl5trv+KYeNZuPgJFG3j4hEFJVAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e4ab23d60ba48ad44e01ffcbfc7c3d5372ee8d9d7c04ab502caffa153976c29","last_reissued_at":"2026-07-05T09:15:26.485043Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:26.485043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TensorGPT: Efficient Compression of Large Language Models based on Tensor-Train Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NA","cs.NE","math.NA"],"primary_cat":"cs.CL","authors_text":"Danilo P. Mandic, Mingxue Xu, Yao Lei Xu","submitted_at":"2023-07-02T09:33:09Z","abstract_excerpt":"High-dimensional token embeddings underpin Large Language Models (LLMs), as they can capture subtle semantic information and significantly enhance the modelling of complex language patterns. However, this high dimensionality also introduces considerable model parameters and prohibitively high model storage and memory requirements, which is particularly unaffordable for low-end devices. Targeting no extra training data and insufficient computation cases, we propose a training-free model compression approach based on the Tensor-Train Decomposition (TTD), whereby each pre-trained token embedding "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.00526","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/2307.00526/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":"2307.00526","created_at":"2026-07-05T09:15:26.485097+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.00526v2","created_at":"2026-07-05T09:15:26.485097+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.00526","created_at":"2026-07-05T09:15:26.485097+00:00"},{"alias_kind":"pith_short_12","alias_value":"FZFLEPLAXJEK","created_at":"2026-07-05T09:15:26.485097+00:00"},{"alias_kind":"pith_short_16","alias_value":"FZFLEPLAXJEK2RHA","created_at":"2026-07-05T09:15:26.485097+00:00"},{"alias_kind":"pith_short_8","alias_value":"FZFLEPLA","created_at":"2026-07-05T09:15:26.485097+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2312.05821","citing_title":"ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":144,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24940","citing_title":"ADE: Adaptive Dictionary Embeddings -- Scaling Multi-Anchor Representations to Large Language Models","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U","json":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U.json","graph_json":"https://pith.science/api/pith-number/FZFLEPLAXJEK2RHAD76L7R6D2U/graph.json","events_json":"https://pith.science/api/pith-number/FZFLEPLAXJEK2RHAD76L7R6D2U/events.json","paper":"https://pith.science/paper/FZFLEPLA"},"agent_actions":{"view_html":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U","download_json":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U.json","view_paper":"https://pith.science/paper/FZFLEPLA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.00526&json=true","fetch_graph":"https://pith.science/api/pith-number/FZFLEPLAXJEK2RHAD76L7R6D2U/graph.json","fetch_events":"https://pith.science/api/pith-number/FZFLEPLAXJEK2RHAD76L7R6D2U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U/action/storage_attestation","attest_author":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U/action/author_attestation","sign_citation":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U/action/citation_signature","submit_replication":"https://pith.science/pith/FZFLEPLAXJEK2RHAD76L7R6D2U/action/replication_record"}},"created_at":"2026-07-05T09:15:26.485097+00:00","updated_at":"2026-07-05T09:15:26.485097+00:00"}