{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:J5OTGOTDYMNXXFOBAF4IF4XENH","short_pith_number":"pith:J5OTGOTD","schema_version":"1.0","canonical_sha256":"4f5d333a63c31b7b95c1017882f2e469d24fa939ee557934db21cfffeab88a1e","source":{"kind":"arxiv","id":"2412.10135","version":2},"attestation_state":"computed","paper":{"title":"ASLoRA: Adaptive Sharing Low-Rank Adaptation Across Layers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junyan Hu, Mengqi Zhang, Pengjie Ren, Xue Xiao, Yao Chen, Zhaochun Ren, Zhumin Chen","submitted_at":"2024-12-13T13:32:13Z","abstract_excerpt":"As large language models (LLMs) grow in size, traditional full fine-tuning becomes increasingly impractical due to its high computational and storage costs. Although popular parameter-efficient fine-tuning methods, such as LoRA, have significantly reduced the number of tunable parameters, there is still room for further optimization. In this work, we propose ASLoRA, a cross-layer parameter-sharing strategy combining global sharing with partial adaptive sharing. Specifically, we share the low-rank matrix A across all layers and adaptively merge matrix B during training. This sharing mechanism n"},"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":"2412.10135","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-13T13:32:13Z","cross_cats_sorted":[],"title_canon_sha256":"c3067682598ffa8020eabaac3f4c0eee3548a94717ca86c811cd245d644cf9b4","abstract_canon_sha256":"68ce64d6ee8993a56efd8805298abeebb33af4169fd1a5dcdcd404089637c7be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:14.774646Z","signature_b64":"D0lpovhIc0Ak78JDoAhwllMU+xJDL194h9xYC3A6Vh1zSOAzXjufLSa3rAOcthO1I+WYB9lwN4kpy/r/H+qPBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f5d333a63c31b7b95c1017882f2e469d24fa939ee557934db21cfffeab88a1e","last_reissued_at":"2026-07-05T09:49:14.774174Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:14.774174Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ASLoRA: Adaptive Sharing Low-Rank Adaptation Across Layers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junyan Hu, Mengqi Zhang, Pengjie Ren, Xue Xiao, Yao Chen, Zhaochun Ren, Zhumin Chen","submitted_at":"2024-12-13T13:32:13Z","abstract_excerpt":"As large language models (LLMs) grow in size, traditional full fine-tuning becomes increasingly impractical due to its high computational and storage costs. Although popular parameter-efficient fine-tuning methods, such as LoRA, have significantly reduced the number of tunable parameters, there is still room for further optimization. In this work, we propose ASLoRA, a cross-layer parameter-sharing strategy combining global sharing with partial adaptive sharing. Specifically, we share the low-rank matrix A across all layers and adaptively merge matrix B during training. This sharing mechanism n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10135","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/2412.10135/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":"2412.10135","created_at":"2026-07-05T09:49:14.774230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10135v2","created_at":"2026-07-05T09:49:14.774230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10135","created_at":"2026-07-05T09:49:14.774230+00:00"},{"alias_kind":"pith_short_12","alias_value":"J5OTGOTDYMNX","created_at":"2026-07-05T09:49:14.774230+00:00"},{"alias_kind":"pith_short_16","alias_value":"J5OTGOTDYMNXXFOB","created_at":"2026-07-05T09:49:14.774230+00:00"},{"alias_kind":"pith_short_8","alias_value":"J5OTGOTD","created_at":"2026-07-05T09:49:14.774230+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/J5OTGOTDYMNXXFOBAF4IF4XENH","json":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH.json","graph_json":"https://pith.science/api/pith-number/J5OTGOTDYMNXXFOBAF4IF4XENH/graph.json","events_json":"https://pith.science/api/pith-number/J5OTGOTDYMNXXFOBAF4IF4XENH/events.json","paper":"https://pith.science/paper/J5OTGOTD"},"agent_actions":{"view_html":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH","download_json":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH.json","view_paper":"https://pith.science/paper/J5OTGOTD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10135&json=true","fetch_graph":"https://pith.science/api/pith-number/J5OTGOTDYMNXXFOBAF4IF4XENH/graph.json","fetch_events":"https://pith.science/api/pith-number/J5OTGOTDYMNXXFOBAF4IF4XENH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH/action/storage_attestation","attest_author":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH/action/author_attestation","sign_citation":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH/action/citation_signature","submit_replication":"https://pith.science/pith/J5OTGOTDYMNXXFOBAF4IF4XENH/action/replication_record"}},"created_at":"2026-07-05T09:49:14.774230+00:00","updated_at":"2026-07-05T09:49:14.774230+00:00"}