{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:HIFH4KG35XT7BW5U3G5WAFMM2F","short_pith_number":"pith:HIFH4KG3","schema_version":"1.0","canonical_sha256":"3a0a7e28dbede7f0dbb4d9bb60158cd1466db58f817b08b2385208ebc72eba4a","source":{"kind":"arxiv","id":"2205.11961","version":2},"attestation_state":"computed","paper":{"title":"ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Prompts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Akari Asai, Hannaneh Hajishirzi, Matthew E. Peters, Mohammadreza Salehi","submitted_at":"2022-05-24T10:48:33Z","abstract_excerpt":"This work introduces a new multi-task, parameter-efficient language model (LM) tuning method that learns to transfer knowledge across different tasks via a mixture of soft prompts-small prefix embedding vectors pre-trained for different tasks. Our method, called ATTEMPT (ATTEntional Mixtures of Prompt Tuning), obtains source prompts as encodings of large-scale source tasks into a small number of parameters and trains an attention module to interpolate the source prompts and a newly initialized target prompt for every instance in the target task. During training, only the target task prompt and"},"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":"2205.11961","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-05-24T10:48:33Z","cross_cats_sorted":[],"title_canon_sha256":"602b6d23113723aa41ed7c8aa4ba6e982047958ced75c312f5a9cc17ef4c7a8c","abstract_canon_sha256":"53c16da4b9551ccf0a21935302f3755cf71306cdfb08f355f68cb3b13a64acff"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:21:25.673794Z","signature_b64":"JhrtbVfInRGewAAc9SoKJH7D9ck5J5f+R+mHMbyQfFSGGJ0MqZQijoNi+7KSHBPKDINGoBveH7OV1JTsNMaWBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a0a7e28dbede7f0dbb4d9bb60158cd1466db58f817b08b2385208ebc72eba4a","last_reissued_at":"2026-07-05T05:21:25.673292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:21:25.673292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Prompts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Akari Asai, Hannaneh Hajishirzi, Matthew E. Peters, Mohammadreza Salehi","submitted_at":"2022-05-24T10:48:33Z","abstract_excerpt":"This work introduces a new multi-task, parameter-efficient language model (LM) tuning method that learns to transfer knowledge across different tasks via a mixture of soft prompts-small prefix embedding vectors pre-trained for different tasks. Our method, called ATTEMPT (ATTEntional Mixtures of Prompt Tuning), obtains source prompts as encodings of large-scale source tasks into a small number of parameters and trains an attention module to interpolate the source prompts and a newly initialized target prompt for every instance in the target task. During training, only the target task prompt and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.11961","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/2205.11961/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":"2205.11961","created_at":"2026-07-05T05:21:25.673359+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.11961v2","created_at":"2026-07-05T05:21:25.673359+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.11961","created_at":"2026-07-05T05:21:25.673359+00:00"},{"alias_kind":"pith_short_12","alias_value":"HIFH4KG35XT7","created_at":"2026-07-05T05:21:25.673359+00:00"},{"alias_kind":"pith_short_16","alias_value":"HIFH4KG35XT7BW5U","created_at":"2026-07-05T05:21:25.673359+00:00"},{"alias_kind":"pith_short_8","alias_value":"HIFH4KG3","created_at":"2026-07-05T05:21:25.673359+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.18124","citing_title":"TLoRA: Task-aware Low Rank Adaptation of Large Language Models","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F","json":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F.json","graph_json":"https://pith.science/api/pith-number/HIFH4KG35XT7BW5U3G5WAFMM2F/graph.json","events_json":"https://pith.science/api/pith-number/HIFH4KG35XT7BW5U3G5WAFMM2F/events.json","paper":"https://pith.science/paper/HIFH4KG3"},"agent_actions":{"view_html":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F","download_json":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F.json","view_paper":"https://pith.science/paper/HIFH4KG3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.11961&json=true","fetch_graph":"https://pith.science/api/pith-number/HIFH4KG35XT7BW5U3G5WAFMM2F/graph.json","fetch_events":"https://pith.science/api/pith-number/HIFH4KG35XT7BW5U3G5WAFMM2F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F/action/storage_attestation","attest_author":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F/action/author_attestation","sign_citation":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F/action/citation_signature","submit_replication":"https://pith.science/pith/HIFH4KG35XT7BW5U3G5WAFMM2F/action/replication_record"}},"created_at":"2026-07-05T05:21:25.673359+00:00","updated_at":"2026-07-05T05:21:25.673359+00:00"}