{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:I2V2MLOMLKFWLT5VXMJVTGEAYL","short_pith_number":"pith:I2V2MLOM","schema_version":"1.0","canonical_sha256":"46aba62dcc5a8b65cfb5bb13599880c2f20653194482fbdd3617f83debe15d16","source":{"kind":"arxiv","id":"2303.17568","version":2},"attestation_state":"computed","paper":{"title":"CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-X","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.LG","authors_text":"Andi Wang, Jie Tang, Lei Shen, Qinkai Zheng, Shan Wang, Teng Su, Xiao Xia, Xu Zou, Yang Li, Yufei Xue, Yuxiao Dong, Zhilin Yang, Zihan Wang","submitted_at":"2023-03-30T17:34:01Z","abstract_excerpt":"Large pre-trained code generation models, such as OpenAI Codex, can generate syntax- and function-correct code, making the coding of programmers more productive and our pursuit of artificial general intelligence closer. In this paper, we introduce CodeGeeX, a multilingual model with 13 billion parameters for code generation. CodeGeeX is pre-trained on 850 billion tokens of 23 programming languages as of June 2022. Our extensive experiments suggest that CodeGeeX outperforms multilingual code models of similar scale for both the tasks of code generation and translation on HumanEval-X. Building u"},"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":"2303.17568","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-03-30T17:34:01Z","cross_cats_sorted":["cs.AI","cs.SE"],"title_canon_sha256":"d036bbf52c33de0f9073eb219bb80c8b43c68799fd27cfcb72e2d24dec3becc3","abstract_canon_sha256":"b34e7443c069705fa13296232b7e461d695fc16cde5f894735bb5e78e13a786e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:41:58.687872Z","signature_b64":"VoY0lt6zXEKki4+vCTWqhk+tlLkIE6XkQ2WjXjIAZc+MuncbOMlGQK5ypR4cDuk9fAfdXl8JvxcfgZuuJyhmAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46aba62dcc5a8b65cfb5bb13599880c2f20653194482fbdd3617f83debe15d16","last_reissued_at":"2026-07-05T08:41:58.687473Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:41:58.687473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-X","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.LG","authors_text":"Andi Wang, Jie Tang, Lei Shen, Qinkai Zheng, Shan Wang, Teng Su, Xiao Xia, Xu Zou, Yang Li, Yufei Xue, Yuxiao Dong, Zhilin Yang, Zihan Wang","submitted_at":"2023-03-30T17:34:01Z","abstract_excerpt":"Large pre-trained code generation models, such as OpenAI Codex, can generate syntax- and function-correct code, making the coding of programmers more productive and our pursuit of artificial general intelligence closer. In this paper, we introduce CodeGeeX, a multilingual model with 13 billion parameters for code generation. CodeGeeX is pre-trained on 850 billion tokens of 23 programming languages as of June 2022. Our extensive experiments suggest that CodeGeeX outperforms multilingual code models of similar scale for both the tasks of code generation and translation on HumanEval-X. Building u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.17568","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/2303.17568/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":"2303.17568","created_at":"2026-07-05T08:41:58.687527+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.17568v2","created_at":"2026-07-05T08:41:58.687527+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.17568","created_at":"2026-07-05T08:41:58.687527+00:00"},{"alias_kind":"pith_short_12","alias_value":"I2V2MLOMLKFW","created_at":"2026-07-05T08:41:58.687527+00:00"},{"alias_kind":"pith_short_16","alias_value":"I2V2MLOMLKFWLT5V","created_at":"2026-07-05T08:41:58.687527+00:00"},{"alias_kind":"pith_short_8","alias_value":"I2V2MLOM","created_at":"2026-07-05T08:41:58.687527+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11863","citing_title":"Enhancing LLM-Based Code Translation with Verified Multi-Semantic Representations","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09059","citing_title":"Evaluating LLM-Generated Code: A Benchmark and Developer Study","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27705","citing_title":"Mitigating Position Bias in Transformers via Layer-Specific Positional Embedding Scaling","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07568","citing_title":"A Systematic Study of Behavioral Cloning for Scientific Data Annotation","ref_index":290,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17074","citing_title":"Surveying GenAI-based Automation in Printed Circuit Board Design and Test","ref_index":122,"is_internal_anchor":false},{"citing_arxiv_id":"2402.01411","citing_title":"CodePori: Large-Scale System for Autonomous Software Development Using Multi-Agent Technology","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2412.04590","citing_title":"Specification-Driven Code Translation Powered by Large Language Models: How Far Are We?","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18073","citing_title":"A-ProS: Towards Reliable Autonomous Programming Through Multi-Model Feedback","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2403.17297","citing_title":"InternLM2 Technical Report","ref_index":205,"is_internal_anchor":false},{"citing_arxiv_id":"2312.13010","citing_title":"AgentCoder: Multi-Agent-based Code Generation with Iterative Testing and Optimisation","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2305.01210","citing_title":"Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09059","citing_title":"Evaluating LLM-Generated Code: A Benchmark and Developer Study","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00382","citing_title":"Social Bias in LLM-Generated Code: Benchmark and Mitigation","ref_index":157,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06161","citing_title":"StarCoder: may the source be with you!","ref_index":125,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06161","citing_title":"StarCoder: may the source be with you!","ref_index":117,"is_internal_anchor":false},{"citing_arxiv_id":"2303.18223","citing_title":"A Survey of Large Language Models","ref_index":94,"is_internal_anchor":false},{"citing_arxiv_id":"2403.07974","citing_title":"LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL","json":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL.json","graph_json":"https://pith.science/api/pith-number/I2V2MLOMLKFWLT5VXMJVTGEAYL/graph.json","events_json":"https://pith.science/api/pith-number/I2V2MLOMLKFWLT5VXMJVTGEAYL/events.json","paper":"https://pith.science/paper/I2V2MLOM"},"agent_actions":{"view_html":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL","download_json":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL.json","view_paper":"https://pith.science/paper/I2V2MLOM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.17568&json=true","fetch_graph":"https://pith.science/api/pith-number/I2V2MLOMLKFWLT5VXMJVTGEAYL/graph.json","fetch_events":"https://pith.science/api/pith-number/I2V2MLOMLKFWLT5VXMJVTGEAYL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL/action/storage_attestation","attest_author":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL/action/author_attestation","sign_citation":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL/action/citation_signature","submit_replication":"https://pith.science/pith/I2V2MLOMLKFWLT5VXMJVTGEAYL/action/replication_record"}},"created_at":"2026-07-05T08:41:58.687527+00:00","updated_at":"2026-07-05T08:41:58.687527+00:00"}