{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:NKPKEKAQ2O6NF2X3RISYDAUBNK","short_pith_number":"pith:NKPKEKAQ","schema_version":"1.0","canonical_sha256":"6a9ea22810d3bcd2eafb8a258182816a8c75c28afbfe61b62d358098c6c88c89","source":{"kind":"arxiv","id":"2007.04973","version":4},"attestation_state":"computed","paper":{"title":"Contrastive Code Representation Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.PL","cs.SE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ajay Jain, Ion Stoica, Joseph E. Gonzalez, Paras Jain, Pieter Abbeel, Tianjun Zhang","submitted_at":"2020-07-09T17:59:06Z","abstract_excerpt":"Recent work learns contextual representations of source code by reconstructing tokens from their context. For downstream semantic understanding tasks like summarizing code in English, these representations should ideally capture program functionality. However, we show that the popular reconstruction-based BERT model is sensitive to source code edits, even when the edits preserve semantics. We propose ContraCode: a contrastive pre-training task that learns code functionality, not form. ContraCode pre-trains a neural network to identify functionally similar variants of a program among many non-e"},"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":"2007.04973","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T17:59:06Z","cross_cats_sorted":["cs.AI","cs.PL","cs.SE","stat.ML"],"title_canon_sha256":"6d26788f1217ce5794558f0ac6e48f220d9c43c68de4f1b081ef91e50f1807b6","abstract_canon_sha256":"9bf56181824cd3796efc24e742ab7776eddb107b01500d2e826e3583d8952fe5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:46:32.171222Z","signature_b64":"SjkU93FQcZlVzdiZfImh1PvBF6EcrvWqFz/t8keeZB2PPvQzcO1QRxbx7einG1ZpkQeKx8e8AporJCfffKaUDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a9ea22810d3bcd2eafb8a258182816a8c75c28afbfe61b62d358098c6c88c89","last_reissued_at":"2026-07-05T03:46:32.170799Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:46:32.170799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive Code Representation Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.PL","cs.SE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ajay Jain, Ion Stoica, Joseph E. Gonzalez, Paras Jain, Pieter Abbeel, Tianjun Zhang","submitted_at":"2020-07-09T17:59:06Z","abstract_excerpt":"Recent work learns contextual representations of source code by reconstructing tokens from their context. For downstream semantic understanding tasks like summarizing code in English, these representations should ideally capture program functionality. However, we show that the popular reconstruction-based BERT model is sensitive to source code edits, even when the edits preserve semantics. We propose ContraCode: a contrastive pre-training task that learns code functionality, not form. ContraCode pre-trains a neural network to identify functionally similar variants of a program among many non-e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.04973","kind":"arxiv","version":4},"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/2007.04973/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":"2007.04973","created_at":"2026-07-05T03:46:32.170868+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.04973v4","created_at":"2026-07-05T03:46:32.170868+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.04973","created_at":"2026-07-05T03:46:32.170868+00:00"},{"alias_kind":"pith_short_12","alias_value":"NKPKEKAQ2O6N","created_at":"2026-07-05T03:46:32.170868+00:00"},{"alias_kind":"pith_short_16","alias_value":"NKPKEKAQ2O6NF2X3","created_at":"2026-07-05T03:46:32.170868+00:00"},{"alias_kind":"pith_short_8","alias_value":"NKPKEKAQ","created_at":"2026-07-05T03:46:32.170868+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK","json":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK.json","graph_json":"https://pith.science/api/pith-number/NKPKEKAQ2O6NF2X3RISYDAUBNK/graph.json","events_json":"https://pith.science/api/pith-number/NKPKEKAQ2O6NF2X3RISYDAUBNK/events.json","paper":"https://pith.science/paper/NKPKEKAQ"},"agent_actions":{"view_html":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK","download_json":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK.json","view_paper":"https://pith.science/paper/NKPKEKAQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.04973&json=true","fetch_graph":"https://pith.science/api/pith-number/NKPKEKAQ2O6NF2X3RISYDAUBNK/graph.json","fetch_events":"https://pith.science/api/pith-number/NKPKEKAQ2O6NF2X3RISYDAUBNK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK/action/storage_attestation","attest_author":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK/action/author_attestation","sign_citation":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK/action/citation_signature","submit_replication":"https://pith.science/pith/NKPKEKAQ2O6NF2X3RISYDAUBNK/action/replication_record"}},"created_at":"2026-07-05T03:46:32.170868+00:00","updated_at":"2026-07-05T03:46:32.170868+00:00"}