{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WRYC567V5JSSKSNYKQM3WKCU4C","short_pith_number":"pith:WRYC567V","schema_version":"1.0","canonical_sha256":"b4702efbf5ea652549b85419bb2854e0b2653d2d93e3ac51a492d56d1afb1216","source":{"kind":"arxiv","id":"2506.03524","version":2},"attestation_state":"computed","paper":{"title":"Seed-Coder: Let the Code Model Curate Data for Itself","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.CL","authors_text":"Anxiang Zhang, Bytedance Seed, Chao Li, Chenguang Xi, Daoguang Zan, Dong Huang, Guanghan Ning, Hanzhi Zhou, Jianchong Chen, Jiaze Chen, Jing Su, Jinhua Zhu, Kaibo Liu, Kai Shen, Liang Xiang, Lixin Dong, Shen Zheng, Shulin Xin, Siyao Liu, Tao Sun, Xia Xiao, Xierui Song, Yetao Bai, Yifan Huang, Yifan Sun, Yonghui Wu, Yuyu Zhang","submitted_at":"2025-06-04T03:17:19Z","abstract_excerpt":"Code data in large language model (LLM) pretraining is recognized crucial not only for code-related tasks but also for enhancing general intelligence of LLMs. Current open-source LLMs often heavily rely on human effort to produce their code pretraining data, such as employing hand-crafted filtering rules tailored to individual programming languages, or using human-annotated data to train quality filters. However, these approaches are inherently limited in scalability, prone to subjective biases, and costly to extend and maintain across diverse programming languages. To address these challenges"},"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":"2506.03524","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T03:17:19Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"ca63399483f0e5c62c42d0f6152d23fe56f15ecd57df2cdf2bfc1bdf399cb5ca","abstract_canon_sha256":"a0e58bd84784c21e5732f6257d2d9566a45593da01543bdfc5d40003444f1edf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:10.059595Z","signature_b64":"Hgdxt/+udiaYraTLBr2X5+aljorW9OaMl8qKz/xawoGl2/BsTjFurnFz5fgCFiav3zqpQ0K6CWOZgaxHzQSnAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4702efbf5ea652549b85419bb2854e0b2653d2d93e3ac51a492d56d1afb1216","last_reissued_at":"2026-07-05T11:16:10.058968Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:10.058968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Seed-Coder: Let the Code Model Curate Data for Itself","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.CL","authors_text":"Anxiang Zhang, Bytedance Seed, Chao Li, Chenguang Xi, Daoguang Zan, Dong Huang, Guanghan Ning, Hanzhi Zhou, Jianchong Chen, Jiaze Chen, Jing Su, Jinhua Zhu, Kaibo Liu, Kai Shen, Liang Xiang, Lixin Dong, Shen Zheng, Shulin Xin, Siyao Liu, Tao Sun, Xia Xiao, Xierui Song, Yetao Bai, Yifan Huang, Yifan Sun, Yonghui Wu, Yuyu Zhang","submitted_at":"2025-06-04T03:17:19Z","abstract_excerpt":"Code data in large language model (LLM) pretraining is recognized crucial not only for code-related tasks but also for enhancing general intelligence of LLMs. Current open-source LLMs often heavily rely on human effort to produce their code pretraining data, such as employing hand-crafted filtering rules tailored to individual programming languages, or using human-annotated data to train quality filters. However, these approaches are inherently limited in scalability, prone to subjective biases, and costly to extend and maintain across diverse programming languages. To address these challenges"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.03524","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/2506.03524/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":"2506.03524","created_at":"2026-07-05T11:16:10.059045+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.03524v2","created_at":"2026-07-05T11:16:10.059045+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.03524","created_at":"2026-07-05T11:16:10.059045+00:00"},{"alias_kind":"pith_short_12","alias_value":"WRYC567V5JSS","created_at":"2026-07-05T11:16:10.059045+00:00"},{"alias_kind":"pith_short_16","alias_value":"WRYC567V5JSSKSNY","created_at":"2026-07-05T11:16:10.059045+00:00"},{"alias_kind":"pith_short_8","alias_value":"WRYC567V","created_at":"2026-07-05T11:16:10.059045+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24901","citing_title":"LLM Evolution as an Industry-Scale Ecosystem: A Lifecycle Perspective on Continual Learning","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18286","citing_title":"CODEBLOCK: Learning to Supervise Code at the Right Granularity","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31268","citing_title":"Mellum2 Technical Report","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2603.23566","citing_title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16322","citing_title":"Steerable Instruction Following Coding Data Synthesis with Actor-Parametric Schema Co-Evolution","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2508.02193","citing_title":"Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03144","citing_title":"InCoder-32B-Thinking: Industrial Code World Model for Thinking","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23509","citing_title":"Uncovering Business Logic Bugs via Semantics-Driven Unit Test Generation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05267","citing_title":"Bridging Generation and Training: A Systematic Review of Quality Issues in LLMs for Code","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19201","citing_title":"Cascaded Code Editing: Large-Small Model Collaboration for Effective and Efficient Code Editing","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12666","citing_title":"From Imitation to Discrimination: Progressive Curriculum Learning for Robust Web Navigation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16625","citing_title":"AdaExplore: Failure-Driven Adaptation and Diversity-Preserving Search for Efficient Kernel Generation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04763","citing_title":"How Does Chunking Affect Retrieval-Augmented Code Completion? A Controlled Empirical Study","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C","json":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C.json","graph_json":"https://pith.science/api/pith-number/WRYC567V5JSSKSNYKQM3WKCU4C/graph.json","events_json":"https://pith.science/api/pith-number/WRYC567V5JSSKSNYKQM3WKCU4C/events.json","paper":"https://pith.science/paper/WRYC567V"},"agent_actions":{"view_html":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C","download_json":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C.json","view_paper":"https://pith.science/paper/WRYC567V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.03524&json=true","fetch_graph":"https://pith.science/api/pith-number/WRYC567V5JSSKSNYKQM3WKCU4C/graph.json","fetch_events":"https://pith.science/api/pith-number/WRYC567V5JSSKSNYKQM3WKCU4C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C/action/storage_attestation","attest_author":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C/action/author_attestation","sign_citation":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C/action/citation_signature","submit_replication":"https://pith.science/pith/WRYC567V5JSSKSNYKQM3WKCU4C/action/replication_record"}},"created_at":"2026-07-05T11:16:10.059045+00:00","updated_at":"2026-07-05T11:16:10.059045+00:00"}