{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ZDAURDERYZ7HE7RTG2DOH475TO","short_pith_number":"pith:ZDAURDER","schema_version":"1.0","canonical_sha256":"c8c1488c91c67e727e333686e3f3fd9bb9b25e1f8ccfb1b50e632f5648510cf5","source":{"kind":"arxiv","id":"2205.12679","version":2},"attestation_state":"computed","paper":{"title":"Self-Guided Noise-Free Data Generation for Efficient Zero-Shot Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hang Xu, Jiacheng Ye, Jiahui Gao, Lingpeng Kong, Renjie Pi, Weizhong Zhang, Xiaodan Liang, Yong Lin, Zhenguo Li, Zhiyong Wu","submitted_at":"2022-05-25T11:38:48Z","abstract_excerpt":"There is a rising interest in further exploring the zero-shot learning potential of large pre-trained language models (PLMs). A new paradigm called data-generation-based zero-shot learning has achieved impressive success. In this paradigm, the synthesized data from the PLM acts as the carrier of knowledge, which is used to train a task-specific model with orders of magnitude fewer parameters than the PLM, achieving both higher performance and efficiency than prompt-based zero-shot learning methods on PLMs. The main hurdle of this approach is that the synthesized data from PLM usually contains "},"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":"2205.12679","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-25T11:38:48Z","cross_cats_sorted":[],"title_canon_sha256":"f8fcbfd4095263cd50933fd9552096f4334bdd59ea75cfcd8305743d5d576806","abstract_canon_sha256":"d0bed374f94bcbb8232347fd897b5a12feefa1d15291dd73f48159428855275b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:45:28.779196Z","signature_b64":"cLr3zJYB0KkFzPXiQkXQjoppeY0nd7O6jY25GlZbbyt7jKj3EJ69nLWB3fUm+IOYP/VIQLtNxCfpggcqimpQDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8c1488c91c67e727e333686e3f3fd9bb9b25e1f8ccfb1b50e632f5648510cf5","last_reissued_at":"2026-07-05T05:45:28.778711Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:45:28.778711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Guided Noise-Free Data Generation for Efficient Zero-Shot Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hang Xu, Jiacheng Ye, Jiahui Gao, Lingpeng Kong, Renjie Pi, Weizhong Zhang, Xiaodan Liang, Yong Lin, Zhenguo Li, Zhiyong Wu","submitted_at":"2022-05-25T11:38:48Z","abstract_excerpt":"There is a rising interest in further exploring the zero-shot learning potential of large pre-trained language models (PLMs). A new paradigm called data-generation-based zero-shot learning has achieved impressive success. In this paradigm, the synthesized data from the PLM acts as the carrier of knowledge, which is used to train a task-specific model with orders of magnitude fewer parameters than the PLM, achieving both higher performance and efficiency than prompt-based zero-shot learning methods on PLMs. The main hurdle of this approach is that the synthesized data from PLM usually contains "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.12679","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/2205.12679/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":"2205.12679","created_at":"2026-07-05T05:45:28.778761+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.12679v2","created_at":"2026-07-05T05:45:28.778761+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.12679","created_at":"2026-07-05T05:45:28.778761+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZDAURDERYZ7H","created_at":"2026-07-05T05:45:28.778761+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZDAURDERYZ7HE7RT","created_at":"2026-07-05T05:45:28.778761+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZDAURDER","created_at":"2026-07-05T05:45:28.778761+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2407.21772","citing_title":"ShieldGemma: Generative AI Content Moderation Based on Gemma","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12335","citing_title":"All in One: A Unified Synthetic Data Pipeline for Multimodal Video Understanding","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO","json":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO.json","graph_json":"https://pith.science/api/pith-number/ZDAURDERYZ7HE7RTG2DOH475TO/graph.json","events_json":"https://pith.science/api/pith-number/ZDAURDERYZ7HE7RTG2DOH475TO/events.json","paper":"https://pith.science/paper/ZDAURDER"},"agent_actions":{"view_html":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO","download_json":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO.json","view_paper":"https://pith.science/paper/ZDAURDER","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.12679&json=true","fetch_graph":"https://pith.science/api/pith-number/ZDAURDERYZ7HE7RTG2DOH475TO/graph.json","fetch_events":"https://pith.science/api/pith-number/ZDAURDERYZ7HE7RTG2DOH475TO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO/action/storage_attestation","attest_author":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO/action/author_attestation","sign_citation":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO/action/citation_signature","submit_replication":"https://pith.science/pith/ZDAURDERYZ7HE7RTG2DOH475TO/action/replication_record"}},"created_at":"2026-07-05T05:45:28.778761+00:00","updated_at":"2026-07-05T05:45:28.778761+00:00"}