{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WCJXKTDVIUS2V6C4F4F6HMYJ7W","short_pith_number":"pith:WCJXKTDV","schema_version":"1.0","canonical_sha256":"b093754c754525aaf85c2f0be3b309fd813cbd51964d57636d7b42cb0a592220","source":{"kind":"arxiv","id":"2211.15533","version":1},"attestation_state":"computed","paper":{"title":"The Stack: 3 TB of permissively licensed source code","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Carlos Mu\\~noz Ferrandis, Chenghao Mou, Denis Kocetkov, Dzmitry Bahdanau, Harm de Vries, Jia Li, Leandro Von Werra, Loubna Ben Allal, Margaret Mitchell, Raymond Li, Sean Hughes, Thomas Wolf, Yacine Jernite","submitted_at":"2022-11-20T18:15:30Z","abstract_excerpt":"Large Language Models (LLMs) play an ever-increasing role in the field of Artificial Intelligence (AI)--not only for natural language processing but also for code understanding and generation. To stimulate open and responsible research on LLMs for code, we introduce The Stack, a 3.1 TB dataset consisting of permissively licensed source code in 30 programming languages. We describe how we collect the full dataset, construct a permissively licensed subset, present a data governance plan, discuss limitations, and show promising results on text2code benchmarks by training 350M-parameter decoders o"},"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":"2211.15533","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-20T18:15:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ab8f0c06ed553547dc89fce0a72730eacaa5cc101ba1bac86b5ad6b066f7154c","abstract_canon_sha256":"d958522c221f116fd919a709cf6ec7cb1a4d8361d09863e4d991daf4835cfff3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:20:05.382771Z","signature_b64":"PNKUEz6rJdD2Z3kLnJNyvNPXRipXeqmyOvy/KaphZkdNhuuDb/lgdd2K0qkoT2ihBVo6lp2CiNixVRp5cqipBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b093754c754525aaf85c2f0be3b309fd813cbd51964d57636d7b42cb0a592220","last_reissued_at":"2026-07-05T05:20:05.382312Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:20:05.382312Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Stack: 3 TB of permissively licensed source code","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Carlos Mu\\~noz Ferrandis, Chenghao Mou, Denis Kocetkov, Dzmitry Bahdanau, Harm de Vries, Jia Li, Leandro Von Werra, Loubna Ben Allal, Margaret Mitchell, Raymond Li, Sean Hughes, Thomas Wolf, Yacine Jernite","submitted_at":"2022-11-20T18:15:30Z","abstract_excerpt":"Large Language Models (LLMs) play an ever-increasing role in the field of Artificial Intelligence (AI)--not only for natural language processing but also for code understanding and generation. To stimulate open and responsible research on LLMs for code, we introduce The Stack, a 3.1 TB dataset consisting of permissively licensed source code in 30 programming languages. We describe how we collect the full dataset, construct a permissively licensed subset, present a data governance plan, discuss limitations, and show promising results on text2code benchmarks by training 350M-parameter decoders o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.15533","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/2211.15533/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":"2211.15533","created_at":"2026-07-05T05:20:05.382387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.15533v1","created_at":"2026-07-05T05:20:05.382387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.15533","created_at":"2026-07-05T05:20:05.382387+00:00"},{"alias_kind":"pith_short_12","alias_value":"WCJXKTDVIUS2","created_at":"2026-07-05T05:20:05.382387+00:00"},{"alias_kind":"pith_short_16","alias_value":"WCJXKTDVIUS2V6C4","created_at":"2026-07-05T05:20:05.382387+00:00"},{"alias_kind":"pith_short_8","alias_value":"WCJXKTDV","created_at":"2026-07-05T05:20:05.382387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":25,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09145","citing_title":"PrivCode++: Latent-Conditioned Differentially Private Code Generation for Comprehensive Guarantees","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08444","citing_title":"When LLMs Invent Rust Crates: An Empirical Study of Hallucination Patterns and Mitigation","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30907","citing_title":"BlueFin: Benchmarking LLM Agents on Financial Spreadsheets","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26144","citing_title":"VISTA: An End-to-End Benchmark for Visual Spec-to-Web-App Coding Agents","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28998","citing_title":"Reward-Free Code Alignment from Pretrained or Fine-Tuned LLM: Unpacking the Trade-offs for Code Generation","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29709","citing_title":"Bash-Commenter: Leveraging Syntax-Aware Preference Optimization to Reinforce Large Language Model for Bash Code Comment Generation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25835","citing_title":"Context-Instrumental Data Distillation for Kubernetes Manifest Generation: Method and Experimental Evaluation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01057","citing_title":"3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2503.22760","citing_title":"Malicious and Unintentional Disclosure Risks in Large Language Models for Code Generation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21543","citing_title":"Provable Joint Decontamination for Benchmarking Multiple Large Language Models","ref_index":78,"is_internal_anchor":false},{"citing_arxiv_id":"2305.16264","citing_title":"Scaling Data-Constrained Language Models","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2502.10517","citing_title":"KernelBench: Can LLMs Write Efficient GPU Kernels?","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2309.05463","citing_title":"Textbooks Are All You Need II: phi-1.5 technical report","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13831","citing_title":"Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2504.05299","citing_title":"SmolVLM: Redefining small and efficient multimodal models","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2502.02737","citing_title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","ref_index":187,"is_internal_anchor":false},{"citing_arxiv_id":"2306.11644","citing_title":"Textbooks Are All You Need","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26923","citing_title":"ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08247","citing_title":"LLM Translation of Compiler Intermediate Representation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.00239","citing_title":"A Taxonomy of Programming Languages for Code Generation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05867","citing_title":"On Fixing Insecure AI-Generated Code through Model Fine-Tuning and Prompting Strategies","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06494","citing_title":"From Token Lists to Graph Motifs: Weisfeiler-Lehman Analysis of Sparse Autoencoder Features","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01188","citing_title":"Compute Optimal Tokenization","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06161","citing_title":"StarCoder: may the source be with you!","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2303.18223","citing_title":"A Survey of Large Language Models","ref_index":177,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W","json":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W.json","graph_json":"https://pith.science/api/pith-number/WCJXKTDVIUS2V6C4F4F6HMYJ7W/graph.json","events_json":"https://pith.science/api/pith-number/WCJXKTDVIUS2V6C4F4F6HMYJ7W/events.json","paper":"https://pith.science/paper/WCJXKTDV"},"agent_actions":{"view_html":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W","download_json":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W.json","view_paper":"https://pith.science/paper/WCJXKTDV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.15533&json=true","fetch_graph":"https://pith.science/api/pith-number/WCJXKTDVIUS2V6C4F4F6HMYJ7W/graph.json","fetch_events":"https://pith.science/api/pith-number/WCJXKTDVIUS2V6C4F4F6HMYJ7W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W/action/storage_attestation","attest_author":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W/action/author_attestation","sign_citation":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W/action/citation_signature","submit_replication":"https://pith.science/pith/WCJXKTDVIUS2V6C4F4F6HMYJ7W/action/replication_record"}},"created_at":"2026-07-05T05:20:05.382387+00:00","updated_at":"2026-07-05T05:20:05.382387+00:00"}