{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XH5KH6DC5UYERISBJFYZUVN4XP","short_pith_number":"pith:XH5KH6DC","schema_version":"1.0","canonical_sha256":"b9faa3f862ed3048a24149719a55bcbbf777728de176ad62b147a48faaf969b6","source":{"kind":"arxiv","id":"2305.18507","version":2},"attestation_state":"computed","paper":{"title":"Code Prompting: a Neural Symbolic Method for Complex Reasoning in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haotong Yang, Muhan Zhang, Yi Hu, Zhouchen Lin","submitted_at":"2023-05-29T15:14:09Z","abstract_excerpt":"Large language models (LLMs) have scaled up to unlock a wide range of complex reasoning tasks with the aid of various prompting methods. However, current prompting methods generate natural language intermediate steps to help reasoning, which can cause imperfect task reduction and confusion. To mitigate such limitations, we explore code prompting, a neural symbolic prompting method with both zero-shot and few-shot versions which triggers code as intermediate steps. We conduct experiments on 7 widely-used benchmarks involving symbolic reasoning and arithmetic reasoning. Code prompting generally "},"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":"2305.18507","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-29T15:14:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5bdf9e804be6274224f6ca3ec6ca3a3dfcacd6619ca7152a5499960687b58473","abstract_canon_sha256":"9f5f593b9234a78cc6ab9e6ab129ac7b4f871453bd9213d24759fe1b9995f99f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:58:10.576671Z","signature_b64":"dbWaysxScLO1CIaddWEopwXWvtGxSMzsVUHtLbq3LxgIYam0o1CkMpHqsfc+18QQ8BrW/zZGBUOgrbbSDg0pBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9faa3f862ed3048a24149719a55bcbbf777728de176ad62b147a48faaf969b6","last_reissued_at":"2026-07-05T06:58:10.576031Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:58:10.576031Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Code Prompting: a Neural Symbolic Method for Complex Reasoning in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haotong Yang, Muhan Zhang, Yi Hu, Zhouchen Lin","submitted_at":"2023-05-29T15:14:09Z","abstract_excerpt":"Large language models (LLMs) have scaled up to unlock a wide range of complex reasoning tasks with the aid of various prompting methods. However, current prompting methods generate natural language intermediate steps to help reasoning, which can cause imperfect task reduction and confusion. To mitigate such limitations, we explore code prompting, a neural symbolic prompting method with both zero-shot and few-shot versions which triggers code as intermediate steps. We conduct experiments on 7 widely-used benchmarks involving symbolic reasoning and arithmetic reasoning. Code prompting generally "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18507","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/2305.18507/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":"2305.18507","created_at":"2026-07-05T06:58:10.576103+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18507v2","created_at":"2026-07-05T06:58:10.576103+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18507","created_at":"2026-07-05T06:58:10.576103+00:00"},{"alias_kind":"pith_short_12","alias_value":"XH5KH6DC5UYE","created_at":"2026-07-05T06:58:10.576103+00:00"},{"alias_kind":"pith_short_16","alias_value":"XH5KH6DC5UYERISB","created_at":"2026-07-05T06:58:10.576103+00:00"},{"alias_kind":"pith_short_8","alias_value":"XH5KH6DC","created_at":"2026-07-05T06:58:10.576103+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.02452","citing_title":"Position: How can Graphs Help Large Language Models?","ref_index":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP","json":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP.json","graph_json":"https://pith.science/api/pith-number/XH5KH6DC5UYERISBJFYZUVN4XP/graph.json","events_json":"https://pith.science/api/pith-number/XH5KH6DC5UYERISBJFYZUVN4XP/events.json","paper":"https://pith.science/paper/XH5KH6DC"},"agent_actions":{"view_html":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP","download_json":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP.json","view_paper":"https://pith.science/paper/XH5KH6DC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18507&json=true","fetch_graph":"https://pith.science/api/pith-number/XH5KH6DC5UYERISBJFYZUVN4XP/graph.json","fetch_events":"https://pith.science/api/pith-number/XH5KH6DC5UYERISBJFYZUVN4XP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP/action/storage_attestation","attest_author":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP/action/author_attestation","sign_citation":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP/action/citation_signature","submit_replication":"https://pith.science/pith/XH5KH6DC5UYERISBJFYZUVN4XP/action/replication_record"}},"created_at":"2026-07-05T06:58:10.576103+00:00","updated_at":"2026-07-05T06:58:10.576103+00:00"}