{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:ZU6ARXUO3EGOOJHSWSVAIK2NQ5","short_pith_number":"pith:ZU6ARXUO","schema_version":"1.0","canonical_sha256":"cd3c08de8ed90ce724f2b4aa042b4d875f12bdd562ce5ea71c6196c28370fa6e","source":{"kind":"arxiv","id":"2006.08084","version":3},"attestation_state":"computed","paper":{"title":"Neural Execution Engines: Learning to Execute Subroutines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","cs.PL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Danai Koutra, Kevin Swersky, Milad Hashemi, Parthasarathy Ranganathan, Yujun Yan","submitted_at":"2020-06-15T01:51:37Z","abstract_excerpt":"A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms. This is evidenced by their inability to generalize to data distributions that are outside of their restricted training sets, namely larger inputs and unseen data. We study these generalization issues at the level of numerical subroutines that comprise common algorithms like sorting, shortest paths, and minimum spanning trees. First, we observe that transformer-based sequence-to-sequence models can learn subroutines lik"},"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":"2006.08084","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-15T01:51:37Z","cross_cats_sorted":["cs.NE","cs.PL","stat.ML"],"title_canon_sha256":"596298fdde1a814cdf066f0cc8ded403aa2cc344a3cc0aee929a6025bcd42eee","abstract_canon_sha256":"efb0db6875efa58e6a03fd17ef3087c1c768b6ce9729ecb2387e5311cd6b9e01"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:45:27.869448Z","signature_b64":"uV3bxBalkA8OGzCVhT3a94mZHMuf3GYRWIZMXn/vMLxjnRQoDz5U9bwXFs10Jrupd3YYCnzL8siG8L5WIhO3BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd3c08de8ed90ce724f2b4aa042b4d875f12bdd562ce5ea71c6196c28370fa6e","last_reissued_at":"2026-07-05T01:45:27.868981Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:45:27.868981Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Execution Engines: Learning to Execute Subroutines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","cs.PL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Danai Koutra, Kevin Swersky, Milad Hashemi, Parthasarathy Ranganathan, Yujun Yan","submitted_at":"2020-06-15T01:51:37Z","abstract_excerpt":"A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms. This is evidenced by their inability to generalize to data distributions that are outside of their restricted training sets, namely larger inputs and unseen data. We study these generalization issues at the level of numerical subroutines that comprise common algorithms like sorting, shortest paths, and minimum spanning trees. First, we observe that transformer-based sequence-to-sequence models can learn subroutines lik"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.08084","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/2006.08084/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":"2006.08084","created_at":"2026-07-05T01:45:27.869040+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.08084v3","created_at":"2026-07-05T01:45:27.869040+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.08084","created_at":"2026-07-05T01:45:27.869040+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZU6ARXUO3EGO","created_at":"2026-07-05T01:45:27.869040+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZU6ARXUO3EGOOJHS","created_at":"2026-07-05T01:45:27.869040+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZU6ARXUO","created_at":"2026-07-05T01:45:27.869040+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2201.11903","citing_title":"Chain-of-Thought Prompting Elicits Reasoning in Large Language Models","ref_index":78,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5","json":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5.json","graph_json":"https://pith.science/api/pith-number/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/graph.json","events_json":"https://pith.science/api/pith-number/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/events.json","paper":"https://pith.science/paper/ZU6ARXUO"},"agent_actions":{"view_html":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5","download_json":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5.json","view_paper":"https://pith.science/paper/ZU6ARXUO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.08084&json=true","fetch_graph":"https://pith.science/api/pith-number/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/graph.json","fetch_events":"https://pith.science/api/pith-number/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/action/storage_attestation","attest_author":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/action/author_attestation","sign_citation":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/action/citation_signature","submit_replication":"https://pith.science/pith/ZU6ARXUO3EGOOJHSWSVAIK2NQ5/action/replication_record"}},"created_at":"2026-07-05T01:45:27.869040+00:00","updated_at":"2026-07-05T01:45:27.869040+00:00"}