{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HUCA6C6TPJHKU2LURUP6FDZFSY","short_pith_number":"pith:HUCA6C6T","schema_version":"1.0","canonical_sha256":"3d040f0bd37a4eaa69748d1fe28f25963dd4cf99b38111ec748a370c265d0abc","source":{"kind":"arxiv","id":"2606.25996","version":1},"attestation_state":"computed","paper":{"title":"Autodata: An agentic data scientist to create high quality synthetic data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Chenxi Whitehouse, Eryk Helenowski, Han Fang, Ilia Kulikov, Jack Lanchantin, Jakob Foerster, Jason Weston, Olga Golovneva, Sainbayar Sukhbaatar, Swarnadeep Saha, Tianhao Wu, Weizhe Yuan, Xian Li, Yixin Nie, Yoram Bachrach","submitted_at":"2026-06-24T16:08:31Z","abstract_excerpt":"We introduce Autodata, a general method that enables AI agents to act as data scientists who build high quality training and evaluation data. We show how to train (meta-optimize) such a data scientist agent, so that it learns to create even stronger data. We describe the overall formulation, and a specific practical implementation, Agentic Self-Instruct. We conduct experiments on computer science research tasks, legal reasoning tasks and reasoning with mathematical objects, where we obtain improved results compared to classical synthetic dataset creation methods. Further, meta-optimizing the d"},"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":"2606.25996","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T16:08:31Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"456c1a9eb2b9676f40db5d340714d12f1060266947e7b716d5fbd322cefa09a1","abstract_canon_sha256":"2c1a4bf7f2c42ddf761b87fe31965f5a7b48b5ec960aaad75e8e611bab997c9c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:18:45.314759Z","signature_b64":"jb5sg7oTSIVpjyk/wvQW8LiNlvVpymEw5JyUN5F7rfZPNKBEGIkaRfx4SjiolE4MeUz8SPrqEWuqolZEwTBaDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d040f0bd37a4eaa69748d1fe28f25963dd4cf99b38111ec748a370c265d0abc","last_reissued_at":"2026-06-25T01:18:45.314404Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:18:45.314404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Autodata: An agentic data scientist to create high quality synthetic data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Chenxi Whitehouse, Eryk Helenowski, Han Fang, Ilia Kulikov, Jack Lanchantin, Jakob Foerster, Jason Weston, Olga Golovneva, Sainbayar Sukhbaatar, Swarnadeep Saha, Tianhao Wu, Weizhe Yuan, Xian Li, Yixin Nie, Yoram Bachrach","submitted_at":"2026-06-24T16:08:31Z","abstract_excerpt":"We introduce Autodata, a general method that enables AI agents to act as data scientists who build high quality training and evaluation data. We show how to train (meta-optimize) such a data scientist agent, so that it learns to create even stronger data. We describe the overall formulation, and a specific practical implementation, Agentic Self-Instruct. We conduct experiments on computer science research tasks, legal reasoning tasks and reasoning with mathematical objects, where we obtain improved results compared to classical synthetic dataset creation methods. Further, meta-optimizing the d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.25996","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/2606.25996/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":"2606.25996","created_at":"2026-06-25T01:18:45.314471+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.25996v1","created_at":"2026-06-25T01:18:45.314471+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.25996","created_at":"2026-06-25T01:18:45.314471+00:00"},{"alias_kind":"pith_short_12","alias_value":"HUCA6C6TPJHK","created_at":"2026-06-25T01:18:45.314471+00:00"},{"alias_kind":"pith_short_16","alias_value":"HUCA6C6TPJHKU2LU","created_at":"2026-06-25T01:18:45.314471+00:00"},{"alias_kind":"pith_short_8","alias_value":"HUCA6C6T","created_at":"2026-06-25T01:18:45.314471+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY","json":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY.json","graph_json":"https://pith.science/api/pith-number/HUCA6C6TPJHKU2LURUP6FDZFSY/graph.json","events_json":"https://pith.science/api/pith-number/HUCA6C6TPJHKU2LURUP6FDZFSY/events.json","paper":"https://pith.science/paper/HUCA6C6T"},"agent_actions":{"view_html":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY","download_json":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY.json","view_paper":"https://pith.science/paper/HUCA6C6T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.25996&json=true","fetch_graph":"https://pith.science/api/pith-number/HUCA6C6TPJHKU2LURUP6FDZFSY/graph.json","fetch_events":"https://pith.science/api/pith-number/HUCA6C6TPJHKU2LURUP6FDZFSY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY/action/storage_attestation","attest_author":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY/action/author_attestation","sign_citation":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY/action/citation_signature","submit_replication":"https://pith.science/pith/HUCA6C6TPJHKU2LURUP6FDZFSY/action/replication_record"}},"created_at":"2026-06-25T01:18:45.314471+00:00","updated_at":"2026-06-25T01:18:45.314471+00:00"}