{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VUE66IBDIB7UGX6IDTXW5KQEO2","short_pith_number":"pith:VUE66IBD","schema_version":"1.0","canonical_sha256":"ad09ef2023407f435fc81cef6eaa0476b7cdc9ac7c546d4de6a95d0885a8e676","source":{"kind":"arxiv","id":"2204.11454","version":2},"attestation_state":"computed","paper":{"title":"Natural Language to Code Translation with Execution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.CL","authors_text":"Daniel Fried, Freda Shi, Luke Zettlemoyer, Marjan Ghazvininejad, Sida I. Wang","submitted_at":"2022-04-25T06:06:08Z","abstract_excerpt":"Generative models of code, pretrained on large corpora of programs, have shown great success in translating natural language to code (Chen et al., 2021; Austin et al., 2021; Li et al., 2022, inter alia). While these models do not explicitly incorporate program semantics (i.e., execution results) during training, they are able to generate correct solutions for many problems. However, choosing a single correct program from a generated set for each problem remains challenging. In this work, we introduce execution result--based minimum Bayes risk decoding (MBR-EXEC) for program selection and show "},"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":"2204.11454","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-04-25T06:06:08Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"3c050bd9b880196437d80b445c6b24fdc8324af6be949414ab004aff1eea8d2a","abstract_canon_sha256":"42de8672ead9fa7453cb60cace4a62bf835e8fe12a4d004041509446f9288fdf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:11.641752Z","signature_b64":"BWI1jdFRtW4WSH5iT99cbuq3BmRLkB037MxKtYdwn4yDUMBZUaDC70YAA36W2zruSIrKhCsLx9B38qBBxnkbBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad09ef2023407f435fc81cef6eaa0476b7cdc9ac7c546d4de6a95d0885a8e676","last_reissued_at":"2026-07-05T05:12:11.641225Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:11.641225Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Natural Language to Code Translation with Execution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.CL","authors_text":"Daniel Fried, Freda Shi, Luke Zettlemoyer, Marjan Ghazvininejad, Sida I. Wang","submitted_at":"2022-04-25T06:06:08Z","abstract_excerpt":"Generative models of code, pretrained on large corpora of programs, have shown great success in translating natural language to code (Chen et al., 2021; Austin et al., 2021; Li et al., 2022, inter alia). While these models do not explicitly incorporate program semantics (i.e., execution results) during training, they are able to generate correct solutions for many problems. However, choosing a single correct program from a generated set for each problem remains challenging. In this work, we introduce execution result--based minimum Bayes risk decoding (MBR-EXEC) for program selection and show "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.11454","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/2204.11454/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":"2204.11454","created_at":"2026-07-05T05:12:11.641287+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.11454v2","created_at":"2026-07-05T05:12:11.641287+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.11454","created_at":"2026-07-05T05:12:11.641287+00:00"},{"alias_kind":"pith_short_12","alias_value":"VUE66IBDIB7U","created_at":"2026-07-05T05:12:11.641287+00:00"},{"alias_kind":"pith_short_16","alias_value":"VUE66IBDIB7UGX6I","created_at":"2026-07-05T05:12:11.641287+00:00"},{"alias_kind":"pith_short_8","alias_value":"VUE66IBD","created_at":"2026-07-05T05:12:11.641287+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2207.10397","citing_title":"CodeT: Code Generation with Generated Tests","ref_index":12,"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":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2","json":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2.json","graph_json":"https://pith.science/api/pith-number/VUE66IBDIB7UGX6IDTXW5KQEO2/graph.json","events_json":"https://pith.science/api/pith-number/VUE66IBDIB7UGX6IDTXW5KQEO2/events.json","paper":"https://pith.science/paper/VUE66IBD"},"agent_actions":{"view_html":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2","download_json":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2.json","view_paper":"https://pith.science/paper/VUE66IBD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.11454&json=true","fetch_graph":"https://pith.science/api/pith-number/VUE66IBDIB7UGX6IDTXW5KQEO2/graph.json","fetch_events":"https://pith.science/api/pith-number/VUE66IBDIB7UGX6IDTXW5KQEO2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2/action/storage_attestation","attest_author":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2/action/author_attestation","sign_citation":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2/action/citation_signature","submit_replication":"https://pith.science/pith/VUE66IBDIB7UGX6IDTXW5KQEO2/action/replication_record"}},"created_at":"2026-07-05T05:12:11.641287+00:00","updated_at":"2026-07-05T05:12:11.641287+00:00"}