{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5QSQTF67FLMSTYPBW65GJLLBO4","short_pith_number":"pith:5QSQTF67","schema_version":"1.0","canonical_sha256":"ec250997df2ad929e1e1b7ba64ad61770cef2870c3f05d6d8e326586939874ad","source":{"kind":"arxiv","id":"2503.17195","version":2},"attestation_state":"computed","paper":{"title":"TreeSynth: Synthesizing Diverse Data from Scratch via Tree-Guided Subspace Partitioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Boyang Xue, Chuan Wu, Jiahui Gao, Jingqi Zhou, Jingwei Dong, Jiyue Jiang, Lingpeng Kong, Pengan Chen, Qintong Li, Sheng Wang","submitted_at":"2025-03-21T14:43:23Z","abstract_excerpt":"Model customization necessitates high-quality and diverse datasets, but acquiring such data remains time-consuming and labor-intensive. Despite the great potential of large language models (LLMs) for data synthesis, current approaches are constrained by limited seed data, model biases, and low-variation prompts, resulting in limited diversity and biased distributions with the increase of data scales. To tackle this challenge, we introduce TREESYNTH, a tree-guided subspace-based data synthesis approach inspired by decision trees. It constructs a spatial partitioning tree to recursively divide a"},"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":"2503.17195","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-21T14:43:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fef9f4a292af7d995031343321cc81ad173089f6bb8ff0778fca7679a5a5defe","abstract_canon_sha256":"5125a1fe9579933e9dea0fe9e7bda44740c24f384b48a2fe8d2c8b2ec1b688ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:42.991438Z","signature_b64":"+4PQ0qdSQyW3u+ilnfIWM8vX8ww6SyAUxsCpfo3rNcgqViHtwTk/lc0/b6SrbK86fUP5E9H6+F2OYEk+LyJbBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec250997df2ad929e1e1b7ba64ad61770cef2870c3f05d6d8e326586939874ad","last_reissued_at":"2026-07-05T11:25:42.990824Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:42.990824Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TreeSynth: Synthesizing Diverse Data from Scratch via Tree-Guided Subspace Partitioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Boyang Xue, Chuan Wu, Jiahui Gao, Jingqi Zhou, Jingwei Dong, Jiyue Jiang, Lingpeng Kong, Pengan Chen, Qintong Li, Sheng Wang","submitted_at":"2025-03-21T14:43:23Z","abstract_excerpt":"Model customization necessitates high-quality and diverse datasets, but acquiring such data remains time-consuming and labor-intensive. Despite the great potential of large language models (LLMs) for data synthesis, current approaches are constrained by limited seed data, model biases, and low-variation prompts, resulting in limited diversity and biased distributions with the increase of data scales. To tackle this challenge, we introduce TREESYNTH, a tree-guided subspace-based data synthesis approach inspired by decision trees. It constructs a spatial partitioning tree to recursively divide a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.17195","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/2503.17195/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":"2503.17195","created_at":"2026-07-05T11:25:42.990908+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.17195v2","created_at":"2026-07-05T11:25:42.990908+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.17195","created_at":"2026-07-05T11:25:42.990908+00:00"},{"alias_kind":"pith_short_12","alias_value":"5QSQTF67FLMS","created_at":"2026-07-05T11:25:42.990908+00:00"},{"alias_kind":"pith_short_16","alias_value":"5QSQTF67FLMSTYPB","created_at":"2026-07-05T11:25:42.990908+00:00"},{"alias_kind":"pith_short_8","alias_value":"5QSQTF67","created_at":"2026-07-05T11:25:42.990908+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17614","citing_title":"Characterizing Model-Native Skills","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4","json":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4.json","graph_json":"https://pith.science/api/pith-number/5QSQTF67FLMSTYPBW65GJLLBO4/graph.json","events_json":"https://pith.science/api/pith-number/5QSQTF67FLMSTYPBW65GJLLBO4/events.json","paper":"https://pith.science/paper/5QSQTF67"},"agent_actions":{"view_html":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4","download_json":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4.json","view_paper":"https://pith.science/paper/5QSQTF67","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.17195&json=true","fetch_graph":"https://pith.science/api/pith-number/5QSQTF67FLMSTYPBW65GJLLBO4/graph.json","fetch_events":"https://pith.science/api/pith-number/5QSQTF67FLMSTYPBW65GJLLBO4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4/action/storage_attestation","attest_author":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4/action/author_attestation","sign_citation":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4/action/citation_signature","submit_replication":"https://pith.science/pith/5QSQTF67FLMSTYPBW65GJLLBO4/action/replication_record"}},"created_at":"2026-07-05T11:25:42.990908+00:00","updated_at":"2026-07-05T11:25:42.990908+00:00"}