{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:537RCFLTRUOWGR43TUHMMMBGMT","short_pith_number":"pith:537RCFLT","schema_version":"1.0","canonical_sha256":"eeff1115738d1d63479b9d0ec6302664db9e47a946f5fbe4813ea5be50c0355a","source":{"kind":"arxiv","id":"2312.10365","version":1},"attestation_state":"computed","paper":{"title":"SPT: Fine-Tuning Transformer-based Language Models Efficiently with Sparsification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Han Yang, James Cheng, Peiqi Yin, Xiao Yan, Yuntao Gui","submitted_at":"2023-12-16T07:44:52Z","abstract_excerpt":"Transformer-based large language models (e.g., BERT and GPT) achieve great success, and fine-tuning, which tunes a pre-trained model on a task-specific dataset, is the standard practice to utilize these models for downstream tasks. However, Transformer fine-tuning has long running time and high memory consumption due to the large size of the models. We propose the SPT system to fine-tune Transformer-based models efficiently by introducing sparsity. We observe that the memory consumption of Transformer mainly comes from storing attention weights for multi-head attention (MHA), and the majority "},"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":"2312.10365","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2023-12-16T07:44:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d513bf66af7f4ab4ca3e8ea23ebd9624aa0862300bdb04c6ac098b6fca11956e","abstract_canon_sha256":"007160deae52ed64f9b8adc76d5ff475762cc5d6164e80533c2a823fd76e70c5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:25:05.561137Z","signature_b64":"SEm4Y4hzODGE7Z+ThTl/FANP4oI7DE9paQr9E/fF5nRpjiS/8Pkw+2NbDBDDBvBdK3sMdlgOCm9bbT2OHZAtCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eeff1115738d1d63479b9d0ec6302664db9e47a946f5fbe4813ea5be50c0355a","last_reissued_at":"2026-07-05T07:25:05.560751Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:25:05.560751Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPT: Fine-Tuning Transformer-based Language Models Efficiently with Sparsification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Han Yang, James Cheng, Peiqi Yin, Xiao Yan, Yuntao Gui","submitted_at":"2023-12-16T07:44:52Z","abstract_excerpt":"Transformer-based large language models (e.g., BERT and GPT) achieve great success, and fine-tuning, which tunes a pre-trained model on a task-specific dataset, is the standard practice to utilize these models for downstream tasks. However, Transformer fine-tuning has long running time and high memory consumption due to the large size of the models. We propose the SPT system to fine-tune Transformer-based models efficiently by introducing sparsity. We observe that the memory consumption of Transformer mainly comes from storing attention weights for multi-head attention (MHA), and the majority "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.10365","kind":"arxiv","version":1},"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/2312.10365/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":"2312.10365","created_at":"2026-07-05T07:25:05.560807+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.10365v1","created_at":"2026-07-05T07:25:05.560807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.10365","created_at":"2026-07-05T07:25:05.560807+00:00"},{"alias_kind":"pith_short_12","alias_value":"537RCFLTRUOW","created_at":"2026-07-05T07:25:05.560807+00:00"},{"alias_kind":"pith_short_16","alias_value":"537RCFLTRUOWGR43","created_at":"2026-07-05T07:25:05.560807+00:00"},{"alias_kind":"pith_short_8","alias_value":"537RCFLT","created_at":"2026-07-05T07:25:05.560807+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/537RCFLTRUOWGR43TUHMMMBGMT","json":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT.json","graph_json":"https://pith.science/api/pith-number/537RCFLTRUOWGR43TUHMMMBGMT/graph.json","events_json":"https://pith.science/api/pith-number/537RCFLTRUOWGR43TUHMMMBGMT/events.json","paper":"https://pith.science/paper/537RCFLT"},"agent_actions":{"view_html":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT","download_json":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT.json","view_paper":"https://pith.science/paper/537RCFLT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.10365&json=true","fetch_graph":"https://pith.science/api/pith-number/537RCFLTRUOWGR43TUHMMMBGMT/graph.json","fetch_events":"https://pith.science/api/pith-number/537RCFLTRUOWGR43TUHMMMBGMT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT/action/storage_attestation","attest_author":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT/action/author_attestation","sign_citation":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT/action/citation_signature","submit_replication":"https://pith.science/pith/537RCFLTRUOWGR43TUHMMMBGMT/action/replication_record"}},"created_at":"2026-07-05T07:25:05.560807+00:00","updated_at":"2026-07-05T07:25:05.560807+00:00"}