{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:NWZ2LUZQPZPVMV62NUPT4I7GKI","short_pith_number":"pith:NWZ2LUZQ","canonical_record":{"source":{"id":"2203.13474","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-25T06:55:15Z","cross_cats_sorted":["cs.CL","cs.PL"],"title_canon_sha256":"9ba27ade9785782edaa81753e0dc8d1830e1e65541a9e8a1d83f1c4511ff9249","abstract_canon_sha256":"d936db27aa7f28c7873b54f880d9ed3a292e1595afe3a75199ea36c7bd3f6dab"},"schema_version":"1.0"},"canonical_sha256":"6db3a5d3307e5f5657da6d1f3e23e65215f3144e17649beb2da19056b17d8538","source":{"kind":"arxiv","id":"2203.13474","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.13474","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"arxiv_version","alias_value":"2203.13474v5","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.13474","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"pith_short_12","alias_value":"NWZ2LUZQPZPV","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"pith_short_16","alias_value":"NWZ2LUZQPZPVMV62","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"pith_short_8","alias_value":"NWZ2LUZQ","created_at":"2026-07-05T05:46:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:NWZ2LUZQPZPVMV62NUPT4I7GKI","target":"record","payload":{"canonical_record":{"source":{"id":"2203.13474","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-25T06:55:15Z","cross_cats_sorted":["cs.CL","cs.PL"],"title_canon_sha256":"9ba27ade9785782edaa81753e0dc8d1830e1e65541a9e8a1d83f1c4511ff9249","abstract_canon_sha256":"d936db27aa7f28c7873b54f880d9ed3a292e1595afe3a75199ea36c7bd3f6dab"},"schema_version":"1.0"},"canonical_sha256":"6db3a5d3307e5f5657da6d1f3e23e65215f3144e17649beb2da19056b17d8538","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:46:14.025883Z","signature_b64":"z7nuamhDPUIj6TLv/ClzJdmP+5Kd35AZ5/o0XdVbzyjnzAWL+dqmSTZguxnb+O3EkF3+PjFAHaTcpXwdaHNjDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6db3a5d3307e5f5657da6d1f3e23e65215f3144e17649beb2da19056b17d8538","last_reissued_at":"2026-07-05T05:46:14.025352Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:46:14.025352Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2203.13474","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:46:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Klq1YRfi9msxW/AxM0X0EKTzvS7ngyRXKRydIZQnpRcasJsxGDbB20TzpkhcpC9EvOGsaTa7S0BUZnmf3fIuDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T14:53:13.079015Z"},"content_sha256":"e415c18c00d0efb1137eedef419a5d575800f1c3c11f536fcfccfcd40a6e799e","schema_version":"1.0","event_id":"sha256:e415c18c00d0efb1137eedef419a5d575800f1c3c11f536fcfccfcd40a6e799e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:NWZ2LUZQPZPVMV62NUPT4I7GKI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"The same problem intent given to CodeGen in multiple turns produces substantially better programs than a single turn on the MTPB benchmark.","cross_cats":["cs.CL","cs.PL"],"primary_cat":"cs.LG","authors_text":"Bo Pang, Caiming Xiong, Erik Nijkamp, Hiroaki Hayashi, Huan Wang, Lifu Tu, Silvio Savarese, Yingbo Zhou","submitted_at":"2022-03-25T06:55:15Z","abstract_excerpt":"Program synthesis strives to generate a computer program as a solution to a given problem specification, expressed with input-output examples or natural language descriptions. The prevalence of large language models advances the state-of-the-art for program synthesis, though limited training resources and data impede open access to such models. To democratize this, we train and release a family of large language models up to 16.1B parameters, called CODEGEN, on natural language and programming language data, and open source the training library JAXFORMER. We show the utility of the trained mod"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our analysis on MTPB shows that the same intent provided to CODEGEN in multi-turn fashion significantly improves program synthesis over that provided as a single turn.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The multi-turn prompts in MTPB accurately capture natural human problem decomposition without introducing selection bias or making the benchmark easier than real-world use cases.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CodeGen models match prior SOTA on HumanEval zero-shot code generation, and multi-turn prompting on the new MTPB benchmark significantly improves synthesis over single-turn inputs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The same problem intent given to CodeGen in multiple turns produces substantially better programs than a single turn on the MTPB benchmark.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a62e9dd77e44d9ae3f1646fec1845d4cb3a2b0969e52520f6ca5741ac7e1341a"},"source":{"id":"2203.13474","kind":"arxiv","version":5},"verdict":{"id":"0616d8ac-1155-4b57-ba6b-dd66dd6d1a1c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T16:59:25.733083Z","strongest_claim":"Our analysis on MTPB shows that the same intent provided to CODEGEN in multi-turn fashion significantly improves program synthesis over that provided as a single turn.","one_line_summary":"CodeGen models match prior SOTA on HumanEval zero-shot code generation, and multi-turn prompting on the new MTPB benchmark significantly improves synthesis over single-turn inputs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The multi-turn prompts in MTPB accurately capture natural human problem decomposition without introducing selection bias or making the benchmark easier than real-world use cases.","pith_extraction_headline":"The same problem intent given to CodeGen in multiple turns produces substantially better programs than a single turn on the MTPB benchmark."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2203.13474/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"},"verdict_id":"0616d8ac-1155-4b57-ba6b-dd66dd6d1a1c"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:46:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7IkH+CNGZ8X50b+Am9QL42dugUYkxqD3fpyvbDW9O6dlzu0HO2Y5sHwMkdKTmEB6bcFhcqd8/lLOebx4Nr0+Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T14:53:13.079488Z"},"content_sha256":"88cb897a6f37b08c29ac0dbeb59b21cb6e916287cbcfc93c6304baba7402e13c","schema_version":"1.0","event_id":"sha256:88cb897a6f37b08c29ac0dbeb59b21cb6e916287cbcfc93c6304baba7402e13c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI/bundle.json","state_url":"https://pith.science/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-10T14:53:13Z","links":{"resolver":"https://pith.science/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI","bundle":"https://pith.science/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI/bundle.json","state":"https://pith.science/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NWZ2LUZQPZPVMV62NUPT4I7GKI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:NWZ2LUZQPZPVMV62NUPT4I7GKI","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d936db27aa7f28c7873b54f880d9ed3a292e1595afe3a75199ea36c7bd3f6dab","cross_cats_sorted":["cs.CL","cs.PL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-25T06:55:15Z","title_canon_sha256":"9ba27ade9785782edaa81753e0dc8d1830e1e65541a9e8a1d83f1c4511ff9249"},"schema_version":"1.0","source":{"id":"2203.13474","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.13474","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"arxiv_version","alias_value":"2203.13474v5","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.13474","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"pith_short_12","alias_value":"NWZ2LUZQPZPV","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"pith_short_16","alias_value":"NWZ2LUZQPZPVMV62","created_at":"2026-07-05T05:46:14Z"},{"alias_kind":"pith_short_8","alias_value":"NWZ2LUZQ","created_at":"2026-07-05T05:46:14Z"}],"graph_snapshots":[{"event_id":"sha256:88cb897a6f37b08c29ac0dbeb59b21cb6e916287cbcfc93c6304baba7402e13c","target":"graph","created_at":"2026-07-05T05:46:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our analysis on MTPB shows that the same intent provided to CODEGEN in multi-turn fashion significantly improves program synthesis over that provided as a single turn."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The multi-turn prompts in MTPB accurately capture natural human problem decomposition without introducing selection bias or making the benchmark easier than real-world use cases."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"CodeGen models match prior SOTA on HumanEval zero-shot code generation, and multi-turn prompting on the new MTPB benchmark significantly improves synthesis over single-turn inputs."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"The same problem intent given to CodeGen in multiple turns produces substantially better programs than a single turn on the MTPB benchmark."}],"snapshot_sha256":"a62e9dd77e44d9ae3f1646fec1845d4cb3a2b0969e52520f6ca5741ac7e1341a"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2203.13474/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Program synthesis strives to generate a computer program as a solution to a given problem specification, expressed with input-output examples or natural language descriptions. The prevalence of large language models advances the state-of-the-art for program synthesis, though limited training resources and data impede open access to such models. To democratize this, we train and release a family of large language models up to 16.1B parameters, called CODEGEN, on natural language and programming language data, and open source the training library JAXFORMER. We show the utility of the trained mod","authors_text":"Bo Pang, Caiming Xiong, Erik Nijkamp, Hiroaki Hayashi, Huan Wang, Lifu Tu, Silvio Savarese, Yingbo Zhou","cross_cats":["cs.CL","cs.PL"],"headline":"The same problem intent given to CodeGen in multiple turns produces substantially better programs than a single turn on the MTPB benchmark.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-25T06:55:15Z","title":"CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.13474","kind":"arxiv","version":5},"verdict":{"created_at":"2026-05-13T16:59:25.733083Z","id":"0616d8ac-1155-4b57-ba6b-dd66dd6d1a1c","model_set":{"reader":"grok-4.3"},"one_line_summary":"CodeGen models match prior SOTA on HumanEval zero-shot code generation, and multi-turn prompting on the new MTPB benchmark significantly improves synthesis over single-turn inputs.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"The same problem intent given to CodeGen in multiple turns produces substantially better programs than a single turn on the MTPB benchmark.","strongest_claim":"Our analysis on MTPB shows that the same intent provided to CODEGEN in multi-turn fashion significantly improves program synthesis over that provided as a single turn.","weakest_assumption":"The multi-turn prompts in MTPB accurately capture natural human problem decomposition without introducing selection bias or making the benchmark easier than real-world use cases."}},"verdict_id":"0616d8ac-1155-4b57-ba6b-dd66dd6d1a1c"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e415c18c00d0efb1137eedef419a5d575800f1c3c11f536fcfccfcd40a6e799e","target":"record","created_at":"2026-07-05T05:46:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d936db27aa7f28c7873b54f880d9ed3a292e1595afe3a75199ea36c7bd3f6dab","cross_cats_sorted":["cs.CL","cs.PL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-25T06:55:15Z","title_canon_sha256":"9ba27ade9785782edaa81753e0dc8d1830e1e65541a9e8a1d83f1c4511ff9249"},"schema_version":"1.0","source":{"id":"2203.13474","kind":"arxiv","version":5}},"canonical_sha256":"6db3a5d3307e5f5657da6d1f3e23e65215f3144e17649beb2da19056b17d8538","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6db3a5d3307e5f5657da6d1f3e23e65215f3144e17649beb2da19056b17d8538","first_computed_at":"2026-07-05T05:46:14.025352Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:46:14.025352Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z7nuamhDPUIj6TLv/ClzJdmP+5Kd35AZ5/o0XdVbzyjnzAWL+dqmSTZguxnb+O3EkF3+PjFAHaTcpXwdaHNjDA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:46:14.025883Z","signed_message":"canonical_sha256_bytes"},"source_id":"2203.13474","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e415c18c00d0efb1137eedef419a5d575800f1c3c11f536fcfccfcd40a6e799e","sha256:88cb897a6f37b08c29ac0dbeb59b21cb6e916287cbcfc93c6304baba7402e13c"],"state_sha256":"c52685aa121451dd36b55235790c846ea695c7818bce1c235399b74941db67bb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BFSiPI1YcpDedNwlfJKlNCIbNUu0+W56WLFJgnWPfS1t/PEfPpBJuDax4XWB4pxroQc1UMNGcrcjzi63tjL3BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-10T14:53:13.081583Z","bundle_sha256":"52a0026adeded9f7bce42bea6e41c59b3794e076e2a10b889ed7426206bd6bbd"}}