{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GEU222O7SXDZYARW4D5YUWENE2","short_pith_number":"pith:GEU222O7","schema_version":"1.0","canonical_sha256":"3129ad69df95c79c0236e0fb8a588d26ba870397125f79dc6000d65ee99f35b0","source":{"kind":"arxiv","id":"2404.14361","version":3},"attestation_state":"computed","paper":{"title":"Better Synthetic Data by Retrieving and Transforming Existing Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Graham Neubig, Ritu Gala, Saumya Gandhi, Tongshuang Wu, Vijay Viswanathan","submitted_at":"2024-04-22T17:15:32Z","abstract_excerpt":"Despite recent advances in large language models, building dependable and deployable NLP models typically requires abundant, high-quality training data. However, task-specific data is not available for many use cases, and manually curating task-specific data is labor-intensive. Recent work has studied prompt-driven synthetic data generation using large language models, but these generated datasets tend to lack complexity and diversity. To address these limitations, we introduce a method, DataTune, to make better use of existing, publicly available datasets to improve automatic dataset generati"},"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":"2404.14361","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-22T17:15:32Z","cross_cats_sorted":[],"title_canon_sha256":"fb7f03ec32330c778ee3e4ab812c4ef034517f1f40f14cfa42eddc239fe446d1","abstract_canon_sha256":"3bc03975386e71e0025fef44ebc22a3607b2c6e38d6d9ec4b92f8709203f953c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:12:50.348011Z","signature_b64":"hVm1WTau9w3Yjq5xhmR8eOcEU+8CgTVgBAybMjqUVZqbUH7N4HGTW93C2o5Z/yx3mBVbXtyqQQ6hx9Rl1WSbAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3129ad69df95c79c0236e0fb8a588d26ba870397125f79dc6000d65ee99f35b0","last_reissued_at":"2026-07-05T08:12:50.347614Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:12:50.347614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Better Synthetic Data by Retrieving and Transforming Existing Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Graham Neubig, Ritu Gala, Saumya Gandhi, Tongshuang Wu, Vijay Viswanathan","submitted_at":"2024-04-22T17:15:32Z","abstract_excerpt":"Despite recent advances in large language models, building dependable and deployable NLP models typically requires abundant, high-quality training data. However, task-specific data is not available for many use cases, and manually curating task-specific data is labor-intensive. Recent work has studied prompt-driven synthetic data generation using large language models, but these generated datasets tend to lack complexity and diversity. To address these limitations, we introduce a method, DataTune, to make better use of existing, publicly available datasets to improve automatic dataset generati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.14361","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/2404.14361/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":"2404.14361","created_at":"2026-07-05T08:12:50.347669+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.14361v3","created_at":"2026-07-05T08:12:50.347669+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.14361","created_at":"2026-07-05T08:12:50.347669+00:00"},{"alias_kind":"pith_short_12","alias_value":"GEU222O7SXDZ","created_at":"2026-07-05T08:12:50.347669+00:00"},{"alias_kind":"pith_short_16","alias_value":"GEU222O7SXDZYARW","created_at":"2026-07-05T08:12:50.347669+00:00"},{"alias_kind":"pith_short_8","alias_value":"GEU222O7","created_at":"2026-07-05T08:12:50.347669+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2406.08464","citing_title":"Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing","ref_index":109,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2","json":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2.json","graph_json":"https://pith.science/api/pith-number/GEU222O7SXDZYARW4D5YUWENE2/graph.json","events_json":"https://pith.science/api/pith-number/GEU222O7SXDZYARW4D5YUWENE2/events.json","paper":"https://pith.science/paper/GEU222O7"},"agent_actions":{"view_html":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2","download_json":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2.json","view_paper":"https://pith.science/paper/GEU222O7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.14361&json=true","fetch_graph":"https://pith.science/api/pith-number/GEU222O7SXDZYARW4D5YUWENE2/graph.json","fetch_events":"https://pith.science/api/pith-number/GEU222O7SXDZYARW4D5YUWENE2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2/action/storage_attestation","attest_author":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2/action/author_attestation","sign_citation":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2/action/citation_signature","submit_replication":"https://pith.science/pith/GEU222O7SXDZYARW4D5YUWENE2/action/replication_record"}},"created_at":"2026-07-05T08:12:50.347669+00:00","updated_at":"2026-07-05T08:12:50.347669+00:00"}