{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:MWHWEQBJ247ZE7P6IZL74J2ZH6","short_pith_number":"pith:MWHWEQBJ","schema_version":"1.0","canonical_sha256":"658f624029d73f927dfe4657fe27593f95306828826889188dafda913bb57dda","source":{"kind":"arxiv","id":"2109.05782","version":2},"attestation_state":"computed","paper":{"title":"Effectiveness of Pre-training for Few-shot Intent Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Albert Y.S. Lam, Guangyuan Shi, Haode Zhang, Jiaxin Chen, Li-Ming Zhan, Xiao-Ming Wu, Yuwei Zhang","submitted_at":"2021-09-13T09:00:09Z","abstract_excerpt":"This paper investigates the effectiveness of pre-training for few-shot intent classification. While existing paradigms commonly further pre-train language models such as BERT on a vast amount of unlabeled corpus, we find it highly effective and efficient to simply fine-tune BERT with a small set of labeled utterances from public datasets. Specifically, fine-tuning BERT with roughly 1,000 labeled data yields a pre-trained model -- IntentBERT, which can easily surpass the performance of existing pre-trained models for few-shot intent classification on novel domains with very different semantics."},"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":"2109.05782","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T09:00:09Z","cross_cats_sorted":[],"title_canon_sha256":"3db34a11fb23f22d27b530c79c7590508a0f533245f879957d7c33ccecd5479b","abstract_canon_sha256":"993c4352fedb37a598e0c601903056aa4d92763a55c7b0778f59aeb7142447f7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:06:51.557697Z","signature_b64":"QapmK4d200NpwAkBdO3ry0IQ67v6/3ScG+3xbvSo0OllXyzjf4ukKGT6LVwOn6Fqpr5A+XtcOpAACfOi/RaVBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"658f624029d73f927dfe4657fe27593f95306828826889188dafda913bb57dda","last_reissued_at":"2026-07-05T09:06:51.557275Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:06:51.557275Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Effectiveness of Pre-training for Few-shot Intent Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Albert Y.S. Lam, Guangyuan Shi, Haode Zhang, Jiaxin Chen, Li-Ming Zhan, Xiao-Ming Wu, Yuwei Zhang","submitted_at":"2021-09-13T09:00:09Z","abstract_excerpt":"This paper investigates the effectiveness of pre-training for few-shot intent classification. While existing paradigms commonly further pre-train language models such as BERT on a vast amount of unlabeled corpus, we find it highly effective and efficient to simply fine-tune BERT with a small set of labeled utterances from public datasets. Specifically, fine-tuning BERT with roughly 1,000 labeled data yields a pre-trained model -- IntentBERT, which can easily surpass the performance of existing pre-trained models for few-shot intent classification on novel domains with very different semantics."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05782","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/2109.05782/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":"2109.05782","created_at":"2026-07-05T09:06:51.557343+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.05782v2","created_at":"2026-07-05T09:06:51.557343+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05782","created_at":"2026-07-05T09:06:51.557343+00:00"},{"alias_kind":"pith_short_12","alias_value":"MWHWEQBJ247Z","created_at":"2026-07-05T09:06:51.557343+00:00"},{"alias_kind":"pith_short_16","alias_value":"MWHWEQBJ247ZE7P6","created_at":"2026-07-05T09:06:51.557343+00:00"},{"alias_kind":"pith_short_8","alias_value":"MWHWEQBJ","created_at":"2026-07-05T09:06:51.557343+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.02682","citing_title":"Hybrid Inspection and Task-Based Access Control in Zero-Trust Agentic AI","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6","json":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6.json","graph_json":"https://pith.science/api/pith-number/MWHWEQBJ247ZE7P6IZL74J2ZH6/graph.json","events_json":"https://pith.science/api/pith-number/MWHWEQBJ247ZE7P6IZL74J2ZH6/events.json","paper":"https://pith.science/paper/MWHWEQBJ"},"agent_actions":{"view_html":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6","download_json":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6.json","view_paper":"https://pith.science/paper/MWHWEQBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.05782&json=true","fetch_graph":"https://pith.science/api/pith-number/MWHWEQBJ247ZE7P6IZL74J2ZH6/graph.json","fetch_events":"https://pith.science/api/pith-number/MWHWEQBJ247ZE7P6IZL74J2ZH6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6/action/storage_attestation","attest_author":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6/action/author_attestation","sign_citation":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6/action/citation_signature","submit_replication":"https://pith.science/pith/MWHWEQBJ247ZE7P6IZL74J2ZH6/action/replication_record"}},"created_at":"2026-07-05T09:06:51.557343+00:00","updated_at":"2026-07-05T09:06:51.557343+00:00"}