{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MF76GHFSMZ26FFW6FQEEEFMDP7","short_pith_number":"pith:MF76GHFS","schema_version":"1.0","canonical_sha256":"617fe31cb26675e296de2c084215837ff236442d99e14949e9b901fc88b52e3c","source":{"kind":"arxiv","id":"2509.07858","version":1},"attestation_state":"computed","paper":{"title":"SCoder: Iterative Self-Distillation for Bootstrapping Small-Scale Data Synthesizers to Empower Code LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Changzhi Zhou, Haomin Fu, Linmei Hu, Luhao Zhang, Mengdi Zhang, Xiancai Chen, Xinyu Zhang, Yang Yang","submitted_at":"2025-09-09T15:38:44Z","abstract_excerpt":"Existing code large language models (LLMs) often rely on large-scale instruction data distilled from proprietary LLMs for fine-tuning, which typically incurs high costs. In this paper, we explore the potential of small-scale open-source LLMs (e.g., 7B) as synthesizers for high-quality code instruction data construction. We first observe that the data synthesis capability of small-scale LLMs can be enhanced by training on a few superior data synthesis samples from proprietary LLMs. Building on this, we propose a novel iterative self-distillation approach to bootstrap small-scale LLMs, transform"},"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":"2509.07858","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-09-09T15:38:44Z","cross_cats_sorted":[],"title_canon_sha256":"6a318263668384c5c94fa652bef04f2be45aff107f0d58f69ccc1b320cd0dd16","abstract_canon_sha256":"e2048e1a358ec9f03f7b3780f7f5ae8fb69868ef535478dd81b58ee164184558"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:41.216137Z","signature_b64":"evT/tjqYGxXWLstiPsaeRHTaW61FhdXQ4uWxrRYNGMZp4ohMa/OmvwVMZ3/lhuBf8W8JNhCh3RL41YOnZFPZDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"617fe31cb26675e296de2c084215837ff236442d99e14949e9b901fc88b52e3c","last_reissued_at":"2026-07-05T12:07:41.215612Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:41.215612Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SCoder: Iterative Self-Distillation for Bootstrapping Small-Scale Data Synthesizers to Empower Code LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Changzhi Zhou, Haomin Fu, Linmei Hu, Luhao Zhang, Mengdi Zhang, Xiancai Chen, Xinyu Zhang, Yang Yang","submitted_at":"2025-09-09T15:38:44Z","abstract_excerpt":"Existing code large language models (LLMs) often rely on large-scale instruction data distilled from proprietary LLMs for fine-tuning, which typically incurs high costs. In this paper, we explore the potential of small-scale open-source LLMs (e.g., 7B) as synthesizers for high-quality code instruction data construction. We first observe that the data synthesis capability of small-scale LLMs can be enhanced by training on a few superior data synthesis samples from proprietary LLMs. Building on this, we propose a novel iterative self-distillation approach to bootstrap small-scale LLMs, transform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.07858","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/2509.07858/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":"2509.07858","created_at":"2026-07-05T12:07:41.215666+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.07858v1","created_at":"2026-07-05T12:07:41.215666+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.07858","created_at":"2026-07-05T12:07:41.215666+00:00"},{"alias_kind":"pith_short_12","alias_value":"MF76GHFSMZ26","created_at":"2026-07-05T12:07:41.215666+00:00"},{"alias_kind":"pith_short_16","alias_value":"MF76GHFSMZ26FFW6","created_at":"2026-07-05T12:07:41.215666+00:00"},{"alias_kind":"pith_short_8","alias_value":"MF76GHFS","created_at":"2026-07-05T12:07:41.215666+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/MF76GHFSMZ26FFW6FQEEEFMDP7","json":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7.json","graph_json":"https://pith.science/api/pith-number/MF76GHFSMZ26FFW6FQEEEFMDP7/graph.json","events_json":"https://pith.science/api/pith-number/MF76GHFSMZ26FFW6FQEEEFMDP7/events.json","paper":"https://pith.science/paper/MF76GHFS"},"agent_actions":{"view_html":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7","download_json":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7.json","view_paper":"https://pith.science/paper/MF76GHFS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.07858&json=true","fetch_graph":"https://pith.science/api/pith-number/MF76GHFSMZ26FFW6FQEEEFMDP7/graph.json","fetch_events":"https://pith.science/api/pith-number/MF76GHFSMZ26FFW6FQEEEFMDP7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7/action/storage_attestation","attest_author":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7/action/author_attestation","sign_citation":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7/action/citation_signature","submit_replication":"https://pith.science/pith/MF76GHFSMZ26FFW6FQEEEFMDP7/action/replication_record"}},"created_at":"2026-07-05T12:07:41.215666+00:00","updated_at":"2026-07-05T12:07:41.215666+00:00"}