{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PNGEKFKAQ34JJMGNU5GCHCUQSD","short_pith_number":"pith:PNGEKFKA","schema_version":"1.0","canonical_sha256":"7b4c45154086f894b0cda74c238a9090eabf5c8edf903fac445dde3d052c6957","source":{"kind":"arxiv","id":"2504.12322","version":2},"attestation_state":"computed","paper":{"title":"A Strategic Coordination Framework of Small LLMs Matches Large LLMs in Data Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Conghui He, Honglin Lin, Jiang Wu, Lijun Wu, Qizhi Pei, Xin Gao, Yu Li, Zinan Tang","submitted_at":"2025-04-11T06:13:43Z","abstract_excerpt":"While data synthesis and distillation are promising strategies to enhance small language models, current approaches heavily rely on Large Language Models (LLMs), which suffer from high computational costs, environmental inefficiency, and potential biases inherited from monolithic architectures. In contrast, smaller LLMs are more accessible and sustainable, but their individual capabilities often fall short in generating high-quality, diverse, and reliable data. Inspired by collaborative human processes (e.g., peer review), we propose a multiple small LLMs involved framework, GRA, that aggregat"},"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":"2504.12322","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-11T06:13:43Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"bd7e68ff178607fffdb167f99ad20ed62fe725a9b5570942130a9a794b6183fc","abstract_canon_sha256":"dfc91b1c16d68beb5d8a9c8148b4b27f674bdf0903c91787034cf4339f25b51e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:37.670297Z","signature_b64":"AyZTgmOYzDat6nLSKruxvQn3F0KPKOWF3Y7PBP95TFTgyqIKTTxiYQqQtYTZYQKkKVCbveO1iUf9wo7SPfLZDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b4c45154086f894b0cda74c238a9090eabf5c8edf903fac445dde3d052c6957","last_reissued_at":"2026-07-05T10:51:37.669799Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:37.669799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Strategic Coordination Framework of Small LLMs Matches Large LLMs in Data Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Conghui He, Honglin Lin, Jiang Wu, Lijun Wu, Qizhi Pei, Xin Gao, Yu Li, Zinan Tang","submitted_at":"2025-04-11T06:13:43Z","abstract_excerpt":"While data synthesis and distillation are promising strategies to enhance small language models, current approaches heavily rely on Large Language Models (LLMs), which suffer from high computational costs, environmental inefficiency, and potential biases inherited from monolithic architectures. In contrast, smaller LLMs are more accessible and sustainable, but their individual capabilities often fall short in generating high-quality, diverse, and reliable data. Inspired by collaborative human processes (e.g., peer review), we propose a multiple small LLMs involved framework, GRA, that aggregat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12322","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/2504.12322/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":"2504.12322","created_at":"2026-07-05T10:51:37.669857+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12322v2","created_at":"2026-07-05T10:51:37.669857+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12322","created_at":"2026-07-05T10:51:37.669857+00:00"},{"alias_kind":"pith_short_12","alias_value":"PNGEKFKAQ34J","created_at":"2026-07-05T10:51:37.669857+00:00"},{"alias_kind":"pith_short_16","alias_value":"PNGEKFKAQ34JJMGN","created_at":"2026-07-05T10:51:37.669857+00:00"},{"alias_kind":"pith_short_8","alias_value":"PNGEKFKA","created_at":"2026-07-05T10:51:37.669857+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22175","citing_title":"StickyInvoc: Rethinking Task Models for High-throughput Workflows in the LLM Era","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12991","citing_title":"Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21166","citing_title":"The Ratchet Effect in Silico: How Interaction Drives Cumulative Intelligence in Large Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12991","citing_title":"Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD","json":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD.json","graph_json":"https://pith.science/api/pith-number/PNGEKFKAQ34JJMGNU5GCHCUQSD/graph.json","events_json":"https://pith.science/api/pith-number/PNGEKFKAQ34JJMGNU5GCHCUQSD/events.json","paper":"https://pith.science/paper/PNGEKFKA"},"agent_actions":{"view_html":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD","download_json":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD.json","view_paper":"https://pith.science/paper/PNGEKFKA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12322&json=true","fetch_graph":"https://pith.science/api/pith-number/PNGEKFKAQ34JJMGNU5GCHCUQSD/graph.json","fetch_events":"https://pith.science/api/pith-number/PNGEKFKAQ34JJMGNU5GCHCUQSD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD/action/storage_attestation","attest_author":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD/action/author_attestation","sign_citation":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD/action/citation_signature","submit_replication":"https://pith.science/pith/PNGEKFKAQ34JJMGNU5GCHCUQSD/action/replication_record"}},"created_at":"2026-07-05T10:51:37.669857+00:00","updated_at":"2026-07-05T10:51:37.669857+00:00"}