{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JGWRKJIY5FR6LRX46DC27LMVIO","short_pith_number":"pith:JGWRKJIY","schema_version":"1.0","canonical_sha256":"49ad152518e963e5c6fcf0c5afad954398fb0fc4b7606d6ffb1254519cc83063","source":{"kind":"arxiv","id":"2210.12587","version":3},"attestation_state":"computed","paper":{"title":"Model ensemble instead of prompt fusion: a sample-specific knowledge transfer method for few-shot prompt tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Chen Xing, Chien-Sheng Wu, Prafulla Kumar Choubey, Xiangyu Peng","submitted_at":"2022-10-23T01:33:16Z","abstract_excerpt":"Prompt tuning approaches, which learn task-specific soft prompts for a downstream task conditioning on frozen pre-trained models, have attracted growing interest due to its parameter efficiency. With large language models and sufficient training data, prompt tuning performs comparably to full-model tuning. However, with limited training samples in few-shot settings, prompt tuning fails to match the performance of full-model fine-tuning. In this work, we focus on improving the few-shot performance of prompt tuning by transferring knowledge from soft prompts of source tasks. Recognizing the good"},"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":"2210.12587","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-10-23T01:33:16Z","cross_cats_sorted":[],"title_canon_sha256":"59fbb5a869bde4ed5d5e6182b0ed98449d40a5193d38dd0e89900adbb4046e25","abstract_canon_sha256":"29288bd5d1e04c4e736939b3cb91fbc9e6b4bd002f9f31656fb91d1c514c1627"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:47:04.469466Z","signature_b64":"4ljO8KDvgEkgqdLDpe0gjyafxcD7fdR4rlQOf/8NUB9lRqZA4I+i586nwE1orhPzPO7y9Mhs6cGPnMVyknAGCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"49ad152518e963e5c6fcf0c5afad954398fb0fc4b7606d6ffb1254519cc83063","last_reissued_at":"2026-07-05T05:47:04.468960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:47:04.468960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model ensemble instead of prompt fusion: a sample-specific knowledge transfer method for few-shot prompt tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Chen Xing, Chien-Sheng Wu, Prafulla Kumar Choubey, Xiangyu Peng","submitted_at":"2022-10-23T01:33:16Z","abstract_excerpt":"Prompt tuning approaches, which learn task-specific soft prompts for a downstream task conditioning on frozen pre-trained models, have attracted growing interest due to its parameter efficiency. With large language models and sufficient training data, prompt tuning performs comparably to full-model tuning. However, with limited training samples in few-shot settings, prompt tuning fails to match the performance of full-model fine-tuning. In this work, we focus on improving the few-shot performance of prompt tuning by transferring knowledge from soft prompts of source tasks. Recognizing the good"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.12587","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/2210.12587/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":"2210.12587","created_at":"2026-07-05T05:47:04.469018+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.12587v3","created_at":"2026-07-05T05:47:04.469018+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.12587","created_at":"2026-07-05T05:47:04.469018+00:00"},{"alias_kind":"pith_short_12","alias_value":"JGWRKJIY5FR6","created_at":"2026-07-05T05:47:04.469018+00:00"},{"alias_kind":"pith_short_16","alias_value":"JGWRKJIY5FR6LRX4","created_at":"2026-07-05T05:47:04.469018+00:00"},{"alias_kind":"pith_short_8","alias_value":"JGWRKJIY","created_at":"2026-07-05T05:47:04.469018+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/JGWRKJIY5FR6LRX46DC27LMVIO","json":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO.json","graph_json":"https://pith.science/api/pith-number/JGWRKJIY5FR6LRX46DC27LMVIO/graph.json","events_json":"https://pith.science/api/pith-number/JGWRKJIY5FR6LRX46DC27LMVIO/events.json","paper":"https://pith.science/paper/JGWRKJIY"},"agent_actions":{"view_html":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO","download_json":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO.json","view_paper":"https://pith.science/paper/JGWRKJIY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.12587&json=true","fetch_graph":"https://pith.science/api/pith-number/JGWRKJIY5FR6LRX46DC27LMVIO/graph.json","fetch_events":"https://pith.science/api/pith-number/JGWRKJIY5FR6LRX46DC27LMVIO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO/action/storage_attestation","attest_author":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO/action/author_attestation","sign_citation":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO/action/citation_signature","submit_replication":"https://pith.science/pith/JGWRKJIY5FR6LRX46DC27LMVIO/action/replication_record"}},"created_at":"2026-07-05T05:47:04.469018+00:00","updated_at":"2026-07-05T05:47:04.469018+00:00"}