{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:22DAKT5PM2Q43HNLZ3GYPTCEJP","short_pith_number":"pith:22DAKT5P","schema_version":"1.0","canonical_sha256":"d686054faf66a1cd9dabcecd87cc444bfa5f3a051e6d8512f0a27a935a50c446","source":{"kind":"arxiv","id":"2306.00477","version":4},"attestation_state":"computed","paper":{"title":"Make Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Baohao Liao, Christof Monz, Shaomu Tan","submitted_at":"2023-06-01T09:26:17Z","abstract_excerpt":"Parameter-efficient fine-tuning (PEFT) of pre-trained language models (PLMs) has emerged as a highly successful approach, with training only a small number of parameters without sacrificing performance and becoming the de-facto learning paradigm with the increasing size of PLMs. However, existing PEFT methods are not memory-efficient, because they still require caching most of the intermediate activations for the gradient calculation, akin to fine-tuning. One effective way to reduce the activation memory is to apply a reversible model, so the intermediate activations are not necessary to be ca"},"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":"2306.00477","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T09:26:17Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3e20f3fba33b501fdb9aaa886cb1298e0abdb2722c73af5d7a9ecb9f67c55367","abstract_canon_sha256":"091988161c866094566eea4cb35e9470fcc8fe4a902f48884a07b168a70cc1d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:29.909088Z","signature_b64":"34Hibfk6A60L0xXNrgy+spQ86VnapQoTlLMULlEzeZ5rL5xPxnFTEiDSthYTEivF/1JyUHopIJh8xdeCgrC5DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d686054faf66a1cd9dabcecd87cc444bfa5f3a051e6d8512f0a27a935a50c446","last_reissued_at":"2026-07-05T07:02:29.908648Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:29.908648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Make Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Baohao Liao, Christof Monz, Shaomu Tan","submitted_at":"2023-06-01T09:26:17Z","abstract_excerpt":"Parameter-efficient fine-tuning (PEFT) of pre-trained language models (PLMs) has emerged as a highly successful approach, with training only a small number of parameters without sacrificing performance and becoming the de-facto learning paradigm with the increasing size of PLMs. However, existing PEFT methods are not memory-efficient, because they still require caching most of the intermediate activations for the gradient calculation, akin to fine-tuning. One effective way to reduce the activation memory is to apply a reversible model, so the intermediate activations are not necessary to be ca"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.00477","kind":"arxiv","version":4},"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/2306.00477/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":"2306.00477","created_at":"2026-07-05T07:02:29.908707+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.00477v4","created_at":"2026-07-05T07:02:29.908707+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.00477","created_at":"2026-07-05T07:02:29.908707+00:00"},{"alias_kind":"pith_short_12","alias_value":"22DAKT5PM2Q4","created_at":"2026-07-05T07:02:29.908707+00:00"},{"alias_kind":"pith_short_16","alias_value":"22DAKT5PM2Q43HNL","created_at":"2026-07-05T07:02:29.908707+00:00"},{"alias_kind":"pith_short_8","alias_value":"22DAKT5P","created_at":"2026-07-05T07:02:29.908707+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2407.17491","citing_title":"Robust Adaptation of Foundation Models with Black-Box Visual Prompting","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":134,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP","json":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP.json","graph_json":"https://pith.science/api/pith-number/22DAKT5PM2Q43HNLZ3GYPTCEJP/graph.json","events_json":"https://pith.science/api/pith-number/22DAKT5PM2Q43HNLZ3GYPTCEJP/events.json","paper":"https://pith.science/paper/22DAKT5P"},"agent_actions":{"view_html":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP","download_json":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP.json","view_paper":"https://pith.science/paper/22DAKT5P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.00477&json=true","fetch_graph":"https://pith.science/api/pith-number/22DAKT5PM2Q43HNLZ3GYPTCEJP/graph.json","fetch_events":"https://pith.science/api/pith-number/22DAKT5PM2Q43HNLZ3GYPTCEJP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP/action/storage_attestation","attest_author":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP/action/author_attestation","sign_citation":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP/action/citation_signature","submit_replication":"https://pith.science/pith/22DAKT5PM2Q43HNLZ3GYPTCEJP/action/replication_record"}},"created_at":"2026-07-05T07:02:29.908707+00:00","updated_at":"2026-07-05T07:02:29.908707+00:00"}