{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:44YD2RP2375CVHYI3DBR476JJY","short_pith_number":"pith:44YD2RP2","schema_version":"1.0","canonical_sha256":"e7303d45fadffa2a9f08d8c31e7fc94e0a708e21e18deb1fe226775168d443b5","source":{"kind":"arxiv","id":"2410.03765","version":1},"attestation_state":"computed","paper":{"title":"Basis Sharing: Cross-Layer Parameter Sharing for Large Language Model Compression","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Bing Li, Grace Li Zhang, Ing-Chao Lin, Jingcun Wang, Yu-Guang Chen","submitted_at":"2024-10-02T14:30:02Z","abstract_excerpt":"Large Language Models (LLMs) have achieved remarkable breakthroughs. However, the huge number of parameters in LLMs require significant amount of memory storage in inference, which prevents their practical deployment in many applications. To reduce memory storage of LLMs, singular value decomposition (SVD) provides a promising solution to approximate weight matrices for compressing LLMs. In this paper, we take a step further to explore parameter sharing across different layers with SVD to achieve more effective compression for LLMs. Specifically, weight matrices in different layers are decompo"},"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":"2410.03765","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-02T14:30:02Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ac3b687472b90bec03ad6f6660adb2ed4626e2a941ca1535a106353ef78be744","abstract_canon_sha256":"9439dc4e82b291097c1b3439c685d355ef6e67d1938c99238bdc606ffa65726e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:24.574773Z","signature_b64":"GjviM9IPXjmyjQ/zHQLTtSqGqoaQNqHSakxegbHCTRH88HJUJ4gUUsKDhIaxgGf691Rdq3v04YZa3hnKoF+PAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7303d45fadffa2a9f08d8c31e7fc94e0a708e21e18deb1fe226775168d443b5","last_reissued_at":"2026-07-05T09:16:24.574313Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:24.574313Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Basis Sharing: Cross-Layer Parameter Sharing for Large Language Model Compression","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Bing Li, Grace Li Zhang, Ing-Chao Lin, Jingcun Wang, Yu-Guang Chen","submitted_at":"2024-10-02T14:30:02Z","abstract_excerpt":"Large Language Models (LLMs) have achieved remarkable breakthroughs. However, the huge number of parameters in LLMs require significant amount of memory storage in inference, which prevents their practical deployment in many applications. To reduce memory storage of LLMs, singular value decomposition (SVD) provides a promising solution to approximate weight matrices for compressing LLMs. In this paper, we take a step further to explore parameter sharing across different layers with SVD to achieve more effective compression for LLMs. Specifically, weight matrices in different layers are decompo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03765","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/2410.03765/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":"2410.03765","created_at":"2026-07-05T09:16:24.574387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.03765v1","created_at":"2026-07-05T09:16:24.574387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03765","created_at":"2026-07-05T09:16:24.574387+00:00"},{"alias_kind":"pith_short_12","alias_value":"44YD2RP2375C","created_at":"2026-07-05T09:16:24.574387+00:00"},{"alias_kind":"pith_short_16","alias_value":"44YD2RP2375CVHYI","created_at":"2026-07-05T09:16:24.574387+00:00"},{"alias_kind":"pith_short_8","alias_value":"44YD2RP2","created_at":"2026-07-05T09:16:24.574387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.22075","citing_title":"CoSpaDi: Compressing LLMs via Calibration-Guided Sparse Dictionary Learning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08314","citing_title":"FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08568","citing_title":"Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13226","citing_title":"KV Packet: Recomputation-Free Context-Independent KV Caching for LLMs","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY","json":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY.json","graph_json":"https://pith.science/api/pith-number/44YD2RP2375CVHYI3DBR476JJY/graph.json","events_json":"https://pith.science/api/pith-number/44YD2RP2375CVHYI3DBR476JJY/events.json","paper":"https://pith.science/paper/44YD2RP2"},"agent_actions":{"view_html":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY","download_json":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY.json","view_paper":"https://pith.science/paper/44YD2RP2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.03765&json=true","fetch_graph":"https://pith.science/api/pith-number/44YD2RP2375CVHYI3DBR476JJY/graph.json","fetch_events":"https://pith.science/api/pith-number/44YD2RP2375CVHYI3DBR476JJY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY/action/storage_attestation","attest_author":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY/action/author_attestation","sign_citation":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY/action/citation_signature","submit_replication":"https://pith.science/pith/44YD2RP2375CVHYI3DBR476JJY/action/replication_record"}},"created_at":"2026-07-05T09:16:24.574387+00:00","updated_at":"2026-07-05T09:16:24.574387+00:00"}