{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6N54X5RZ7CEUMJWJ56CRLVT3RY","short_pith_number":"pith:6N54X5RZ","schema_version":"1.0","canonical_sha256":"f37bcbf639f8894626c9ef8515d67b8e2c17abcb5b60e2de343709cb02360f28","source":{"kind":"arxiv","id":"2210.17451","version":2},"attestation_state":"computed","paper":{"title":"AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ahmed Hassan Awadallah, Jianfeng Gao, Jing Gao, Sahaj Agarwal, Subhabrata Mukherjee, XiaoDong Liu, Yaqing Wang","submitted_at":"2022-10-31T16:23:36Z","abstract_excerpt":"Standard fine-tuning of large pre-trained language models (PLMs) for downstream tasks requires updating hundreds of millions to billions of parameters, and storing a large copy of the PLM weights for every task resulting in increased cost for storing, sharing and serving the models. To address this, parameter-efficient fine-tuning (PEFT) techniques were introduced where small trainable components are injected in the PLM and updated during fine-tuning. We propose AdaMix as a general PEFT method that tunes a mixture of adaptation modules -- given the underlying PEFT method of choice -- introduce"},"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":"2210.17451","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-31T16:23:36Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0271ca7fd3f49b5b6e4c4e6a49528473db3d9325e5fe58cfab2a2efb5c9b2800","abstract_canon_sha256":"0543650958ae285e6a9a058a33fede91ab832b978bda222d75901e4f2b3914f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:32.278793Z","signature_b64":"At9svZHmeCNnjOi9GbPUhXd8LxWaHH2SgaNBGkKxDfX88Jnmsn0zAjvsbA4gmtsLpgOyGk83LBvf+GbH9nGgCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f37bcbf639f8894626c9ef8515d67b8e2c17abcb5b60e2de343709cb02360f28","last_reissued_at":"2026-07-05T05:12:32.278378Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:32.278378Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ahmed Hassan Awadallah, Jianfeng Gao, Jing Gao, Sahaj Agarwal, Subhabrata Mukherjee, XiaoDong Liu, Yaqing Wang","submitted_at":"2022-10-31T16:23:36Z","abstract_excerpt":"Standard fine-tuning of large pre-trained language models (PLMs) for downstream tasks requires updating hundreds of millions to billions of parameters, and storing a large copy of the PLM weights for every task resulting in increased cost for storing, sharing and serving the models. To address this, parameter-efficient fine-tuning (PEFT) techniques were introduced where small trainable components are injected in the PLM and updated during fine-tuning. We propose AdaMix as a general PEFT method that tunes a mixture of adaptation modules -- given the underlying PEFT method of choice -- introduce"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.17451","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/2210.17451/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":"2210.17451","created_at":"2026-07-05T05:12:32.278440+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.17451v2","created_at":"2026-07-05T05:12:32.278440+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.17451","created_at":"2026-07-05T05:12:32.278440+00:00"},{"alias_kind":"pith_short_12","alias_value":"6N54X5RZ7CEU","created_at":"2026-07-05T05:12:32.278440+00:00"},{"alias_kind":"pith_short_16","alias_value":"6N54X5RZ7CEUMJWJ","created_at":"2026-07-05T05:12:32.278440+00:00"},{"alias_kind":"pith_short_8","alias_value":"6N54X5RZ","created_at":"2026-07-05T05:12:32.278440+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.15831","citing_title":"Efficient and Private: Memorisation under differentially private parameter-efficient fine-tuning in language models","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY","json":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY.json","graph_json":"https://pith.science/api/pith-number/6N54X5RZ7CEUMJWJ56CRLVT3RY/graph.json","events_json":"https://pith.science/api/pith-number/6N54X5RZ7CEUMJWJ56CRLVT3RY/events.json","paper":"https://pith.science/paper/6N54X5RZ"},"agent_actions":{"view_html":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY","download_json":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY.json","view_paper":"https://pith.science/paper/6N54X5RZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.17451&json=true","fetch_graph":"https://pith.science/api/pith-number/6N54X5RZ7CEUMJWJ56CRLVT3RY/graph.json","fetch_events":"https://pith.science/api/pith-number/6N54X5RZ7CEUMJWJ56CRLVT3RY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY/action/storage_attestation","attest_author":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY/action/author_attestation","sign_citation":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY/action/citation_signature","submit_replication":"https://pith.science/pith/6N54X5RZ7CEUMJWJ56CRLVT3RY/action/replication_record"}},"created_at":"2026-07-05T05:12:32.278440+00:00","updated_at":"2026-07-05T05:12:32.278440+00:00"}