{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PADO257E7HEXRZFSTZV7E7PQ46","short_pith_number":"pith:PADO257E","schema_version":"1.0","canonical_sha256":"7806ed77e4f9c978e4b29e6bf27df0e7ae8e58fadace29c28d31157a4fafb531","source":{"kind":"arxiv","id":"2406.15734","version":2},"attestation_state":"computed","paper":{"title":"RankAdaptor: Hierarchical Rank Allocation for Efficient Fine-Tuning Pruned LLMs via Performance Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Changhai Zhou, Hongguang Li, Lining Yang, Shijie Han, Xu Cheng, Yibin Wang, Yuhua Zhou","submitted_at":"2024-06-22T04:52:58Z","abstract_excerpt":"The efficient compression of large language models (LLMs) has become increasingly popular. However, recovering the performance of compressed LLMs remains a major challenge. The current practice in LLM compression entails the implementation of structural pruning, complemented by a recovery phase that leverages the Low-Rank Adaptation (LoRA) algorithm. Structural pruning's uneven modification of model architecture, coupled with standard LoRA's fixed configuration allocation across layers in an online pipeline, leads to suboptimal performance in various downstream tasks for pruned models. To addr"},"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":"2406.15734","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-22T04:52:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c324e006421595142b3686eea1e51ca14cef240ed2cd702b9ee87e206c2c1321","abstract_canon_sha256":"4c9579c526504854b97fa775bcda99b998fc37ef0da4a2f40641dbe03707790f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:07.108292Z","signature_b64":"7f7vPfwBAjFAuYurqtIKCyDJqoyB3OL2cAx6eVZ+r42Nw5Wc1oaQhqG7qzSEcf5GiyBu2kA6eR4xXHAFMcviAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7806ed77e4f9c978e4b29e6bf27df0e7ae8e58fadace29c28d31157a4fafb531","last_reissued_at":"2026-07-05T09:49:07.107781Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:07.107781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RankAdaptor: Hierarchical Rank Allocation for Efficient Fine-Tuning Pruned LLMs via Performance Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Changhai Zhou, Hongguang Li, Lining Yang, Shijie Han, Xu Cheng, Yibin Wang, Yuhua Zhou","submitted_at":"2024-06-22T04:52:58Z","abstract_excerpt":"The efficient compression of large language models (LLMs) has become increasingly popular. However, recovering the performance of compressed LLMs remains a major challenge. The current practice in LLM compression entails the implementation of structural pruning, complemented by a recovery phase that leverages the Low-Rank Adaptation (LoRA) algorithm. Structural pruning's uneven modification of model architecture, coupled with standard LoRA's fixed configuration allocation across layers in an online pipeline, leads to suboptimal performance in various downstream tasks for pruned models. To addr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.15734","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/2406.15734/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":"2406.15734","created_at":"2026-07-05T09:49:07.107841+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.15734v2","created_at":"2026-07-05T09:49:07.107841+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.15734","created_at":"2026-07-05T09:49:07.107841+00:00"},{"alias_kind":"pith_short_12","alias_value":"PADO257E7HEX","created_at":"2026-07-05T09:49:07.107841+00:00"},{"alias_kind":"pith_short_16","alias_value":"PADO257E7HEXRZFS","created_at":"2026-07-05T09:49:07.107841+00:00"},{"alias_kind":"pith_short_8","alias_value":"PADO257E","created_at":"2026-07-05T09:49:07.107841+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07500","citing_title":"Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46","json":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46.json","graph_json":"https://pith.science/api/pith-number/PADO257E7HEXRZFSTZV7E7PQ46/graph.json","events_json":"https://pith.science/api/pith-number/PADO257E7HEXRZFSTZV7E7PQ46/events.json","paper":"https://pith.science/paper/PADO257E"},"agent_actions":{"view_html":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46","download_json":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46.json","view_paper":"https://pith.science/paper/PADO257E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.15734&json=true","fetch_graph":"https://pith.science/api/pith-number/PADO257E7HEXRZFSTZV7E7PQ46/graph.json","fetch_events":"https://pith.science/api/pith-number/PADO257E7HEXRZFSTZV7E7PQ46/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46/action/storage_attestation","attest_author":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46/action/author_attestation","sign_citation":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46/action/citation_signature","submit_replication":"https://pith.science/pith/PADO257E7HEXRZFSTZV7E7PQ46/action/replication_record"}},"created_at":"2026-07-05T09:49:07.107841+00:00","updated_at":"2026-07-05T09:49:07.107841+00:00"}