{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:IWHWVQCFXWKZRLR7UCDQ7POGEH","short_pith_number":"pith:IWHWVQCF","schema_version":"1.0","canonical_sha256":"458f6ac045bd9598ae3fa0870fbdc621cab9530c27d23dba8f88b2bd31bc6953","source":{"kind":"arxiv","id":"2305.18403","version":5},"attestation_state":"computed","paper":{"title":"LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Bohan Zhuang, Chunhua Shen, Hao Chen, Linlin Ou, Mingyang Zhang, Xinyi Yu, Zhen Yang","submitted_at":"2023-05-28T15:15:48Z","abstract_excerpt":"Large Language Models (LLMs), such as LLaMA and T5, have shown exceptional performance across various tasks through fine-tuning. Although low-rank adaption (LoRA) has emerged to cheaply fine-tune these LLMs on downstream tasks, their deployment is still hindered by the vast model scale and computational costs. Post-training model pruning offers a way to compress LLMs. However, the current pruning methods designed for LLMs are not compatible with LoRA. This is due to their utilization of unstructured pruning on LLMs, impeding the merging of LoRA weights, or their dependence on the gradients of "},"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":"2305.18403","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-28T15:15:48Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"0b50abaa90c7b620aa7551bb7dec547aff3e86cf7b59f990af91d7656c42ef08","abstract_canon_sha256":"4c1da2d9578072eda512625b450799eba58c5fd2c27cbb36c454f6014ce338db"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:52:50.439210Z","signature_b64":"GPV8dEdMEjpDeIrFPjy8D68vMnsS0LPSlxUq1s23RmfgVL9Wvfr3oF4DnQyXSRIQVDo+67ZWKjaW1xaxSARkCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"458f6ac045bd9598ae3fa0870fbdc621cab9530c27d23dba8f88b2bd31bc6953","last_reissued_at":"2026-07-05T08:52:50.438761Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:52:50.438761Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Bohan Zhuang, Chunhua Shen, Hao Chen, Linlin Ou, Mingyang Zhang, Xinyi Yu, Zhen Yang","submitted_at":"2023-05-28T15:15:48Z","abstract_excerpt":"Large Language Models (LLMs), such as LLaMA and T5, have shown exceptional performance across various tasks through fine-tuning. Although low-rank adaption (LoRA) has emerged to cheaply fine-tune these LLMs on downstream tasks, their deployment is still hindered by the vast model scale and computational costs. Post-training model pruning offers a way to compress LLMs. However, the current pruning methods designed for LLMs are not compatible with LoRA. This is due to their utilization of unstructured pruning on LLMs, impeding the merging of LoRA weights, or their dependence on the gradients of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18403","kind":"arxiv","version":5},"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/2305.18403/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":"2305.18403","created_at":"2026-07-05T08:52:50.438821+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18403v5","created_at":"2026-07-05T08:52:50.438821+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18403","created_at":"2026-07-05T08:52:50.438821+00:00"},{"alias_kind":"pith_short_12","alias_value":"IWHWVQCFXWKZ","created_at":"2026-07-05T08:52:50.438821+00:00"},{"alias_kind":"pith_short_16","alias_value":"IWHWVQCFXWKZRLR7","created_at":"2026-07-05T08:52:50.438821+00:00"},{"alias_kind":"pith_short_8","alias_value":"IWHWVQCF","created_at":"2026-07-05T08:52:50.438821+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07557","citing_title":"PALS: Percentile-Aware Layerwise Sparsity for LLM Pruning","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01412","citing_title":"GPTQ-intrinsic LoRA: A Near-optimal Algorithm for Low-precision Quantization with Low-rank Adaptation","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2503.08223","citing_title":"Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices","ref_index":169,"is_internal_anchor":false},{"citing_arxiv_id":"2505.17138","citing_title":"RAP: Runtime Adaptive Pruning for LLM Inference","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2506.12876","citing_title":"MaskPro: Linear-Space Probabilistic Learning for Strict (N:M)-Sparsity on LLMs","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":120,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH","json":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH.json","graph_json":"https://pith.science/api/pith-number/IWHWVQCFXWKZRLR7UCDQ7POGEH/graph.json","events_json":"https://pith.science/api/pith-number/IWHWVQCFXWKZRLR7UCDQ7POGEH/events.json","paper":"https://pith.science/paper/IWHWVQCF"},"agent_actions":{"view_html":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH","download_json":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH.json","view_paper":"https://pith.science/paper/IWHWVQCF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18403&json=true","fetch_graph":"https://pith.science/api/pith-number/IWHWVQCFXWKZRLR7UCDQ7POGEH/graph.json","fetch_events":"https://pith.science/api/pith-number/IWHWVQCFXWKZRLR7UCDQ7POGEH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH/action/storage_attestation","attest_author":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH/action/author_attestation","sign_citation":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH/action/citation_signature","submit_replication":"https://pith.science/pith/IWHWVQCFXWKZRLR7UCDQ7POGEH/action/replication_record"}},"created_at":"2026-07-05T08:52:50.438821+00:00","updated_at":"2026-07-05T08:52:50.438821+00:00"}