{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BAD6YCLINZKUUJGOWRLFXECPSQ","short_pith_number":"pith:BAD6YCLI","schema_version":"1.0","canonical_sha256":"0807ec09686e554a24ceb4565b904f9436424848ffca148de419fd6e510940d1","source":{"kind":"arxiv","id":"2305.10703","version":1},"attestation_state":"computed","paper":{"title":"ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chao Zhang, Jiaming Shen, Rongzhi Zhang, Yuchen Zhuang, Yue Yu, Yu Meng","submitted_at":"2023-05-18T04:30:09Z","abstract_excerpt":"With the development of large language models (LLMs), zero-shot learning has attracted much attention for various NLP tasks. Different from prior works that generate training data with billion-scale natural language generation (NLG) models, we propose a retrieval-enhanced framework to create training data from a general-domain unlabeled corpus. To realize this, we first conduct contrastive pretraining to learn an unsupervised dense retriever for extracting the most relevant documents using class-descriptive verbalizers. We then further propose two simple strategies, namely Verbalizer Augmentat"},"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":"2305.10703","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-18T04:30:09Z","cross_cats_sorted":["cs.IR","cs.LG"],"title_canon_sha256":"d1a4509b1ef95c51b48c4a16a2f7699b767a5a2febc61e7d24617ae078485ae1","abstract_canon_sha256":"38239d6e66a290d95ec1d4fdad106ad7d285b8d39a0ecb82bf2a39c7ff0ced3f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:11:25.717097Z","signature_b64":"zLUHAmz29/bFn3ULPfPx0hb8i50OGGmGTf396KksMvecHNFTxPwOrSehdKIaQmjRO4oCzyomp5M+7BL61pKYDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0807ec09686e554a24ceb4565b904f9436424848ffca148de419fd6e510940d1","last_reissued_at":"2026-07-05T06:11:25.716687Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:11:25.716687Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chao Zhang, Jiaming Shen, Rongzhi Zhang, Yuchen Zhuang, Yue Yu, Yu Meng","submitted_at":"2023-05-18T04:30:09Z","abstract_excerpt":"With the development of large language models (LLMs), zero-shot learning has attracted much attention for various NLP tasks. Different from prior works that generate training data with billion-scale natural language generation (NLG) models, we propose a retrieval-enhanced framework to create training data from a general-domain unlabeled corpus. To realize this, we first conduct contrastive pretraining to learn an unsupervised dense retriever for extracting the most relevant documents using class-descriptive verbalizers. We then further propose two simple strategies, namely Verbalizer Augmentat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.10703","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/2305.10703/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":"2305.10703","created_at":"2026-07-05T06:11:25.716752+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.10703v1","created_at":"2026-07-05T06:11:25.716752+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.10703","created_at":"2026-07-05T06:11:25.716752+00:00"},{"alias_kind":"pith_short_12","alias_value":"BAD6YCLINZKU","created_at":"2026-07-05T06:11:25.716752+00:00"},{"alias_kind":"pith_short_16","alias_value":"BAD6YCLINZKUUJGO","created_at":"2026-07-05T06:11:25.716752+00:00"},{"alias_kind":"pith_short_8","alias_value":"BAD6YCLI","created_at":"2026-07-05T06:11:25.716752+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.04172","citing_title":"Does Prompt Design Impact Quality of Data Imputation by LLMs?","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ","json":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ.json","graph_json":"https://pith.science/api/pith-number/BAD6YCLINZKUUJGOWRLFXECPSQ/graph.json","events_json":"https://pith.science/api/pith-number/BAD6YCLINZKUUJGOWRLFXECPSQ/events.json","paper":"https://pith.science/paper/BAD6YCLI"},"agent_actions":{"view_html":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ","download_json":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ.json","view_paper":"https://pith.science/paper/BAD6YCLI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.10703&json=true","fetch_graph":"https://pith.science/api/pith-number/BAD6YCLINZKUUJGOWRLFXECPSQ/graph.json","fetch_events":"https://pith.science/api/pith-number/BAD6YCLINZKUUJGOWRLFXECPSQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ/action/storage_attestation","attest_author":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ/action/author_attestation","sign_citation":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ/action/citation_signature","submit_replication":"https://pith.science/pith/BAD6YCLINZKUUJGOWRLFXECPSQ/action/replication_record"}},"created_at":"2026-07-05T06:11:25.716752+00:00","updated_at":"2026-07-05T06:11:25.716752+00:00"}