{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5547JMA64S6IC6LFMNRGDU55HR","short_pith_number":"pith:5547JMA6","schema_version":"1.0","canonical_sha256":"ef79f4b01ee4bc817965636261d3bd3c5c8783cc2006ab7857e869bf2305e665","source":{"kind":"arxiv","id":"2510.00192","version":3},"attestation_state":"computed","paper":{"title":"Train Large, Deploy Compact: Structured Compression for Compact Low-Rank Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Cong Xie, Lingzhou Xue, Tiantian Fan, Xin Yu, XunMei Liu, Zhi Zhang","submitted_at":"2025-09-30T19:10:35Z","abstract_excerpt":"Low-rank adaptation (LoRA) has become a widely used paradigm for parameter-efficient fine-tuning of large language models, yet its representational capacity often lags behind full fine-tuning. Within the context of LoRA, a key open question is how to obtain expressive low-rank adapters from over-parameterized spaces. We propose \\textit{PrunedLoRA}, a new framework that leverages structured pruning to obtain highly representative low-rank adapters from an over-parameterized initialization. Unlike prior approaches that impose a fixed low-rank budget, PrunedLoRA dynamically prunes less important "},"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":"2510.00192","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-30T19:10:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"85a9c8a6eb86885897ee033a53f9b08123cb770cb294b78c6796661a66853cca","abstract_canon_sha256":"a1c88ff7424d23738ee33552768319761cec54e0405a38b3fcb04591a7651ce1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef79f4b01ee4bc817965636261d3bd3c5c8783cc2006ab7857e869bf2305e665","last_reissued_at":"2026-07-30T01:16:17.127756Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:16:17.127756Z"},"graph_snapshot":{"paper":{"title":"Train Large, Deploy Compact: Structured Compression for Compact Low-Rank Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Cong Xie, Lingzhou Xue, Tiantian Fan, Xin Yu, XunMei Liu, Zhi Zhang","submitted_at":"2025-09-30T19:10:35Z","abstract_excerpt":"Low-rank adaptation (LoRA) has become a widely used paradigm for parameter-efficient fine-tuning of large language models, yet its representational capacity often lags behind full fine-tuning. Within the context of LoRA, a key open question is how to obtain expressive low-rank adapters from over-parameterized spaces. We propose \\textit{PrunedLoRA}, a new framework that leverages structured pruning to obtain highly representative low-rank adapters from an over-parameterized initialization. Unlike prior approaches that impose a fixed low-rank budget, PrunedLoRA dynamically prunes less important "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.00192","kind":"arxiv","version":3},"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/2510.00192/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":"2510.00192","created_at":"2026-07-30T01:16:17.132936+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.00192v3","created_at":"2026-07-30T01:16:17.132936+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.00192","created_at":"2026-07-30T01:16:17.132936+00:00"},{"alias_kind":"pith_short_12","alias_value":"5547JMA64S6I","created_at":"2026-07-30T01:16:17.132936+00:00"},{"alias_kind":"pith_short_16","alias_value":"5547JMA64S6IC6LF","created_at":"2026-07-30T01:16:17.132936+00:00"},{"alias_kind":"pith_short_8","alias_value":"5547JMA6","created_at":"2026-07-30T01:16:17.132936+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR","json":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR.json","graph_json":"https://pith.science/api/pith-number/5547JMA64S6IC6LFMNRGDU55HR/graph.json","events_json":"https://pith.science/api/pith-number/5547JMA64S6IC6LFMNRGDU55HR/events.json","paper":"https://pith.science/paper/5547JMA6"},"agent_actions":{"view_html":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR","download_json":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR.json","view_paper":"https://pith.science/paper/5547JMA6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.00192&json=true","fetch_graph":"https://pith.science/api/pith-number/5547JMA64S6IC6LFMNRGDU55HR/graph.json","fetch_events":"https://pith.science/api/pith-number/5547JMA64S6IC6LFMNRGDU55HR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR/action/storage_attestation","attest_author":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR/action/author_attestation","sign_citation":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR/action/citation_signature","submit_replication":"https://pith.science/pith/5547JMA64S6IC6LFMNRGDU55HR/action/replication_record"}},"created_at":"2026-07-30T01:16:17.132936+00:00","updated_at":"2026-07-30T01:16:17.132936+00:00"}