{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VP6Y2GZHRPMNGKDSMJHZ6KNJCZ","short_pith_number":"pith:VP6Y2GZH","schema_version":"1.0","canonical_sha256":"abfd8d1b278bd8d32872624f9f29a91664f44d39a68bc64b58b490ce9b8cbba6","source":{"kind":"arxiv","id":"2207.05987","version":3},"attestation_state":"computed","paper":{"title":"DocPrompting: Generating Code by Retrieving the Docs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.CL","authors_text":"Frank F. Xu, Graham Neubig, Shuyan Zhou, Uri Alon, Zhengbao Jiang, Zhiruo Wang","submitted_at":"2022-07-13T06:47:51Z","abstract_excerpt":"Publicly available source-code libraries are continuously growing and changing. This makes it impossible for models of code to keep current with all available APIs by simply training these models on existing code repositories. Thus, existing models inherently cannot generalize to using unseen functions and libraries, because these would never appear in the training data. In contrast, when human programmers use functions and libraries for the first time, they frequently refer to textual resources such as code manuals and documentation, to explore and understand the available functionality. Insp"},"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":"2207.05987","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-07-13T06:47:51Z","cross_cats_sorted":["cs.AI","cs.SE"],"title_canon_sha256":"9a8c0eacd380fb18dc38c2ad42c6ea0374c6219cbb07956ed3ff1dfb8f9d4dce","abstract_canon_sha256":"8f58ab1da8d2688637de0bef0b66f17341dd3963bfa512540c559e693ff28292"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:43:13.673363Z","signature_b64":"TtDpZD1KxrE+DA6Lbol1YfOKp5C90TcSruJZ3NyPgss7NUjpgbLTUdz9QV6UfqAwxSEy95hAeQpKc3VmL+DKAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abfd8d1b278bd8d32872624f9f29a91664f44d39a68bc64b58b490ce9b8cbba6","last_reissued_at":"2026-07-05T05:43:13.672906Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:43:13.672906Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DocPrompting: Generating Code by Retrieving the Docs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.CL","authors_text":"Frank F. Xu, Graham Neubig, Shuyan Zhou, Uri Alon, Zhengbao Jiang, Zhiruo Wang","submitted_at":"2022-07-13T06:47:51Z","abstract_excerpt":"Publicly available source-code libraries are continuously growing and changing. This makes it impossible for models of code to keep current with all available APIs by simply training these models on existing code repositories. Thus, existing models inherently cannot generalize to using unseen functions and libraries, because these would never appear in the training data. In contrast, when human programmers use functions and libraries for the first time, they frequently refer to textual resources such as code manuals and documentation, to explore and understand the available functionality. Insp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.05987","kind":"arxiv","version":3},"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/2207.05987/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":"2207.05987","created_at":"2026-07-05T05:43:13.672965+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.05987v3","created_at":"2026-07-05T05:43:13.672965+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.05987","created_at":"2026-07-05T05:43:13.672965+00:00"},{"alias_kind":"pith_short_12","alias_value":"VP6Y2GZHRPMN","created_at":"2026-07-05T05:43:13.672965+00:00"},{"alias_kind":"pith_short_16","alias_value":"VP6Y2GZHRPMNGKDS","created_at":"2026-07-05T05:43:13.672965+00:00"},{"alias_kind":"pith_short_8","alias_value":"VP6Y2GZH","created_at":"2026-07-05T05:43:13.672965+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.08715","citing_title":"VeriInteresting: An Empirical Study of Model Prompt Interactions in Verilog Code Generation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14503","citing_title":"Not All RAGs Are Created Equal: A Component-Wise Empirical Study for Software Engineering Tasks","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12159","citing_title":"ALGOGEN: Tool-Generated Verifiable Traces for Reliable Algorithm Visualization","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19826","citing_title":"Co-Located Tests, Better AI Code: How Test Syntax Structure Affects Foundation Model Code Generation","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01698","citing_title":"BIM Information Extraction Through LLM-based Adaptive Exploration","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ","json":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ.json","graph_json":"https://pith.science/api/pith-number/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/graph.json","events_json":"https://pith.science/api/pith-number/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/events.json","paper":"https://pith.science/paper/VP6Y2GZH"},"agent_actions":{"view_html":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ","download_json":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ.json","view_paper":"https://pith.science/paper/VP6Y2GZH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.05987&json=true","fetch_graph":"https://pith.science/api/pith-number/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/graph.json","fetch_events":"https://pith.science/api/pith-number/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/action/storage_attestation","attest_author":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/action/author_attestation","sign_citation":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/action/citation_signature","submit_replication":"https://pith.science/pith/VP6Y2GZHRPMNGKDSMJHZ6KNJCZ/action/replication_record"}},"created_at":"2026-07-05T05:43:13.672965+00:00","updated_at":"2026-07-05T05:43:13.672965+00:00"}