{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:2ZKRIA7TC7E7F24H36F57EOBX2","short_pith_number":"pith:2ZKRIA7T","schema_version":"1.0","canonical_sha256":"d6551403f317c9f2eb87df8bdf91c1beace8c738b6b07682fe4e086e6e282ea1","source":{"kind":"arxiv","id":"2104.14690","version":1},"attestation_state":"computed","paper":{"title":"Entailment as Few-Shot Learner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Han Fang, Hanzi Mao, Hao Ma, Madian Khabsa, Sinong Wang","submitted_at":"2021-04-29T22:52:26Z","abstract_excerpt":"Large pre-trained language models (LMs) have demonstrated remarkable ability as few-shot learners. However, their success hinges largely on scaling model parameters to a degree that makes it challenging to train and serve. In this paper, we propose a new approach, named as EFL, that can turn small LMs into better few-shot learners. The key idea of this approach is to reformulate potential NLP task into an entailment one, and then fine-tune the model with as little as 8 examples. We further demonstrate our proposed method can be: (i) naturally combined with an unsupervised contrastive learning-"},"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":"2104.14690","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-29T22:52:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"29e9d508509dfc5c75f6077ae62107147c48077c88d33346678ae6c78eb8977c","abstract_canon_sha256":"3b30fe2178e3a11e79d81e0e58fd730283aec5dfce5f6d308781f30746368d1e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:36:25.373259Z","signature_b64":"kt4sHJGlMPuRUidN6WRJO8KcpRsdtiAM2gbDn7iPemcZbpsOYvKOhccC9AULSXzVBvvGwyU4pPly7BZUZ42mDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6551403f317c9f2eb87df8bdf91c1beace8c738b6b07682fe4e086e6e282ea1","last_reissued_at":"2026-07-05T02:36:25.372860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:36:25.372860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Entailment as Few-Shot Learner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Han Fang, Hanzi Mao, Hao Ma, Madian Khabsa, Sinong Wang","submitted_at":"2021-04-29T22:52:26Z","abstract_excerpt":"Large pre-trained language models (LMs) have demonstrated remarkable ability as few-shot learners. However, their success hinges largely on scaling model parameters to a degree that makes it challenging to train and serve. In this paper, we propose a new approach, named as EFL, that can turn small LMs into better few-shot learners. The key idea of this approach is to reformulate potential NLP task into an entailment one, and then fine-tune the model with as little as 8 examples. We further demonstrate our proposed method can be: (i) naturally combined with an unsupervised contrastive learning-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.14690","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/2104.14690/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":"2104.14690","created_at":"2026-07-05T02:36:25.372915+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.14690v1","created_at":"2026-07-05T02:36:25.372915+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.14690","created_at":"2026-07-05T02:36:25.372915+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ZKRIA7TC7E7","created_at":"2026-07-05T02:36:25.372915+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ZKRIA7TC7E7F24H","created_at":"2026-07-05T02:36:25.372915+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ZKRIA7T","created_at":"2026-07-05T02:36:25.372915+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2110.08207","citing_title":"Multitask Prompted Training Enables Zero-Shot Task Generalization","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2302.09664","citing_title":"Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21469","citing_title":"Cross-Domain Data Selection and Augmentation for Automatic Compliance Detection","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2","json":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2.json","graph_json":"https://pith.science/api/pith-number/2ZKRIA7TC7E7F24H36F57EOBX2/graph.json","events_json":"https://pith.science/api/pith-number/2ZKRIA7TC7E7F24H36F57EOBX2/events.json","paper":"https://pith.science/paper/2ZKRIA7T"},"agent_actions":{"view_html":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2","download_json":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2.json","view_paper":"https://pith.science/paper/2ZKRIA7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.14690&json=true","fetch_graph":"https://pith.science/api/pith-number/2ZKRIA7TC7E7F24H36F57EOBX2/graph.json","fetch_events":"https://pith.science/api/pith-number/2ZKRIA7TC7E7F24H36F57EOBX2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2/action/storage_attestation","attest_author":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2/action/author_attestation","sign_citation":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2/action/citation_signature","submit_replication":"https://pith.science/pith/2ZKRIA7TC7E7F24H36F57EOBX2/action/replication_record"}},"created_at":"2026-07-05T02:36:25.372915+00:00","updated_at":"2026-07-05T02:36:25.372915+00:00"}