{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3CM3A443BIJK2B452VY7CMQ4XK","short_pith_number":"pith:3CM3A443","schema_version":"1.0","canonical_sha256":"d899b0739b0a12ad079dd571f1321cbabebd34bcfc1d65de93494fe116a639de","source":{"kind":"arxiv","id":"2407.20584","version":3},"attestation_state":"computed","paper":{"title":"Pruning Large Language Models with Semi-Structural Adaptive Sparse Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guohao Jian, Jianfei Chen, Jun Zhu, Weiyu Huang, Yuezhou Hu","submitted_at":"2024-07-30T06:33:44Z","abstract_excerpt":"The remarkable success of Large Language Models (LLMs) relies heavily on their substantial scale, which poses significant challenges during model deployment in terms of latency and memory consumption. Recently, numerous studies have attempted to compress LLMs using one-shot pruning methods. However, these methods often suffer from considerable performance degradation on complex language understanding tasks, raising concerns about the feasibility of pruning in LLMs. To address this issue, we propose Adaptive Sparse Trainer (AST), a novel and efficient retraining framework tailored for semi-stru"},"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":"2407.20584","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-30T06:33:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"64ae67cf497a20b5211c17fea180dc520784f3fc6ad842e5ef6347e67d81be5e","abstract_canon_sha256":"0e97dc2764c76ab73eb70dc47ba8f77bedb31844cec64fd84b3ec5c71c130db4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:50:50.548895Z","signature_b64":"5a718QZLYV+oChPHlyQ42BeVK9DA2ub5LMAU3zMZx3nqHZAe6wDLOOsRByn03R24Q9/vtnQ+KAimziHgoWNvCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d899b0739b0a12ad079dd571f1321cbabebd34bcfc1d65de93494fe116a639de","last_reissued_at":"2026-07-05T09:50:50.548275Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:50:50.548275Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pruning Large Language Models with Semi-Structural Adaptive Sparse Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guohao Jian, Jianfei Chen, Jun Zhu, Weiyu Huang, Yuezhou Hu","submitted_at":"2024-07-30T06:33:44Z","abstract_excerpt":"The remarkable success of Large Language Models (LLMs) relies heavily on their substantial scale, which poses significant challenges during model deployment in terms of latency and memory consumption. Recently, numerous studies have attempted to compress LLMs using one-shot pruning methods. However, these methods often suffer from considerable performance degradation on complex language understanding tasks, raising concerns about the feasibility of pruning in LLMs. To address this issue, we propose Adaptive Sparse Trainer (AST), a novel and efficient retraining framework tailored for semi-stru"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.20584","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/2407.20584/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":"2407.20584","created_at":"2026-07-05T09:50:50.548365+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.20584v3","created_at":"2026-07-05T09:50:50.548365+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.20584","created_at":"2026-07-05T09:50:50.548365+00:00"},{"alias_kind":"pith_short_12","alias_value":"3CM3A443BIJK","created_at":"2026-07-05T09:50:50.548365+00:00"},{"alias_kind":"pith_short_16","alias_value":"3CM3A443BIJK2B45","created_at":"2026-07-05T09:50:50.548365+00:00"},{"alias_kind":"pith_short_8","alias_value":"3CM3A443","created_at":"2026-07-05T09:50:50.548365+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.03337","citing_title":"Mitigating Non-IID Drift in Zeroth-Order Federated LLM Fine-Tuning with Transferable Sparsity","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK","json":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK.json","graph_json":"https://pith.science/api/pith-number/3CM3A443BIJK2B452VY7CMQ4XK/graph.json","events_json":"https://pith.science/api/pith-number/3CM3A443BIJK2B452VY7CMQ4XK/events.json","paper":"https://pith.science/paper/3CM3A443"},"agent_actions":{"view_html":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK","download_json":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK.json","view_paper":"https://pith.science/paper/3CM3A443","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.20584&json=true","fetch_graph":"https://pith.science/api/pith-number/3CM3A443BIJK2B452VY7CMQ4XK/graph.json","fetch_events":"https://pith.science/api/pith-number/3CM3A443BIJK2B452VY7CMQ4XK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK/action/storage_attestation","attest_author":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK/action/author_attestation","sign_citation":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK/action/citation_signature","submit_replication":"https://pith.science/pith/3CM3A443BIJK2B452VY7CMQ4XK/action/replication_record"}},"created_at":"2026-07-05T09:50:50.548365+00:00","updated_at":"2026-07-05T09:50:50.548365+00:00"}