{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QHAQY5GRILLRGHYJEZHNMKQNAJ","short_pith_number":"pith:QHAQY5GR","schema_version":"1.0","canonical_sha256":"81c10c74d142d7131f09264ed62a0d027cec3bd108cd387dfece20f013f83f39","source":{"kind":"arxiv","id":"2505.00582","version":2},"attestation_state":"computed","paper":{"title":"Block Circulant Adapter for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Meiqi Wang, Siyu Liao, Xinyu Ding, Zhongfeng Wang","submitted_at":"2025-05-01T15:14:32Z","abstract_excerpt":"Fine-tuning large language models (LLMs) is difficult due to their huge model size. Recent Fourier domain-based methods show potential for reducing fine-tuning costs. We propose a block circulant matrix-based fine-tuning method with a stable training heuristic to leverage the properties of circulant matrices and one-dimensional Fourier transforms to reduce storage and computation costs. Experiments show that our method uses $14\\times$ less number of parameters than VeRA, $16\\times$ smaller than LoRA and $32\\times$ less FLOPs than FourierFT, while maintaining close or better task performance. O"},"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":"2505.00582","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-01T15:14:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5933bd99328452b86d4fb96c1614ab864ab86d995562cdd484db442d2cce87cf","abstract_canon_sha256":"53b73b96bc9f32d77e64c441564a98760fb296264493cd4a9af9d447e15610a1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:37:19.148425Z","signature_b64":"m8lDBZiRyD3RSyXhCF+1VcDnmLn2vszjDOCemCx3tEbjuw9oZddesZfTlTFYTmWtS3x+FWS9sMVYEULoewcMCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81c10c74d142d7131f09264ed62a0d027cec3bd108cd387dfece20f013f83f39","last_reissued_at":"2026-07-05T11:37:19.147437Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:37:19.147437Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Block Circulant Adapter for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Meiqi Wang, Siyu Liao, Xinyu Ding, Zhongfeng Wang","submitted_at":"2025-05-01T15:14:32Z","abstract_excerpt":"Fine-tuning large language models (LLMs) is difficult due to their huge model size. Recent Fourier domain-based methods show potential for reducing fine-tuning costs. We propose a block circulant matrix-based fine-tuning method with a stable training heuristic to leverage the properties of circulant matrices and one-dimensional Fourier transforms to reduce storage and computation costs. Experiments show that our method uses $14\\times$ less number of parameters than VeRA, $16\\times$ smaller than LoRA and $32\\times$ less FLOPs than FourierFT, while maintaining close or better task performance. O"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00582","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/2505.00582/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":"2505.00582","created_at":"2026-07-05T11:37:19.147497+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.00582v2","created_at":"2026-07-05T11:37:19.147497+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00582","created_at":"2026-07-05T11:37:19.147497+00:00"},{"alias_kind":"pith_short_12","alias_value":"QHAQY5GRILLR","created_at":"2026-07-05T11:37:19.147497+00:00"},{"alias_kind":"pith_short_16","alias_value":"QHAQY5GRILLRGHYJ","created_at":"2026-07-05T11:37:19.147497+00:00"},{"alias_kind":"pith_short_8","alias_value":"QHAQY5GR","created_at":"2026-07-05T11:37:19.147497+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16443","citing_title":"Two-Valued Symmetric Circulant Matrices: Applications in Deep Learning","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ","json":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ.json","graph_json":"https://pith.science/api/pith-number/QHAQY5GRILLRGHYJEZHNMKQNAJ/graph.json","events_json":"https://pith.science/api/pith-number/QHAQY5GRILLRGHYJEZHNMKQNAJ/events.json","paper":"https://pith.science/paper/QHAQY5GR"},"agent_actions":{"view_html":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ","download_json":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ.json","view_paper":"https://pith.science/paper/QHAQY5GR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.00582&json=true","fetch_graph":"https://pith.science/api/pith-number/QHAQY5GRILLRGHYJEZHNMKQNAJ/graph.json","fetch_events":"https://pith.science/api/pith-number/QHAQY5GRILLRGHYJEZHNMKQNAJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ/action/storage_attestation","attest_author":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ/action/author_attestation","sign_citation":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ/action/citation_signature","submit_replication":"https://pith.science/pith/QHAQY5GRILLRGHYJEZHNMKQNAJ/action/replication_record"}},"created_at":"2026-07-05T11:37:19.147497+00:00","updated_at":"2026-07-05T11:37:19.147497+00:00"}