{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:ZRXX6SDQAA6WRDMQH7KQJZMAGH","short_pith_number":"pith:ZRXX6SDQ","schema_version":"1.0","canonical_sha256":"cc6f7f4870003d688d903fd504e58031c7cc8ce136b9b70853e55019a35388c5","source":{"kind":"arxiv","id":"1911.01497","version":3},"attestation_state":"computed","paper":{"title":"On Compositionality in Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Florian Metze, Vaibhav Kumar, Vikas Raunak","submitted_at":"2019-11-04T21:31:36Z","abstract_excerpt":"We investigate two specific manifestations of compositionality in Neural Machine Translation (NMT) : (1) Productivity - the ability of the model to extend its predictions beyond the observed length in training data and (2) Systematicity - the ability of the model to systematically recombine known parts and rules. We evaluate a standard Sequence to Sequence model on tests designed to assess these two properties in NMT. We quantitatively demonstrate that inadequate temporal processing, in the form of poor encoder representations is a bottleneck for both Productivity and Systematicity. We propose"},"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":"1911.01497","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-11-04T21:31:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b9aa569808b0e374f00ef37ed902943d1a679051c344ab9e8e5ea84a0c878605","abstract_canon_sha256":"5f2dcfd876c0ed13b61e01768b2ddb336f0979f20e4ce5140b73bebaa9556051"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:26:11.434271Z","signature_b64":"A0PVIILYJ6CKQOAmpQl3YhF5SmmSS//9zr2P0vLct9dZszY5uY3ZE84WZ/WTMZv8KBp5rEpEl37GbuQ7rH9LBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc6f7f4870003d688d903fd504e58031c7cc8ce136b9b70853e55019a35388c5","last_reissued_at":"2026-07-05T00:26:11.433830Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:26:11.433830Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Compositionality in Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Florian Metze, Vaibhav Kumar, Vikas Raunak","submitted_at":"2019-11-04T21:31:36Z","abstract_excerpt":"We investigate two specific manifestations of compositionality in Neural Machine Translation (NMT) : (1) Productivity - the ability of the model to extend its predictions beyond the observed length in training data and (2) Systematicity - the ability of the model to systematically recombine known parts and rules. We evaluate a standard Sequence to Sequence model on tests designed to assess these two properties in NMT. We quantitatively demonstrate that inadequate temporal processing, in the form of poor encoder representations is a bottleneck for both Productivity and Systematicity. We propose"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.01497","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/1911.01497/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":"1911.01497","created_at":"2026-07-05T00:26:11.433897+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.01497v3","created_at":"2026-07-05T00:26:11.433897+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.01497","created_at":"2026-07-05T00:26:11.433897+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZRXX6SDQAA6W","created_at":"2026-07-05T00:26:11.433897+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZRXX6SDQAA6WRDMQ","created_at":"2026-07-05T00:26:11.433897+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZRXX6SDQ","created_at":"2026-07-05T00:26:11.433897+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/ZRXX6SDQAA6WRDMQH7KQJZMAGH","json":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH.json","graph_json":"https://pith.science/api/pith-number/ZRXX6SDQAA6WRDMQH7KQJZMAGH/graph.json","events_json":"https://pith.science/api/pith-number/ZRXX6SDQAA6WRDMQH7KQJZMAGH/events.json","paper":"https://pith.science/paper/ZRXX6SDQ"},"agent_actions":{"view_html":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH","download_json":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH.json","view_paper":"https://pith.science/paper/ZRXX6SDQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.01497&json=true","fetch_graph":"https://pith.science/api/pith-number/ZRXX6SDQAA6WRDMQH7KQJZMAGH/graph.json","fetch_events":"https://pith.science/api/pith-number/ZRXX6SDQAA6WRDMQH7KQJZMAGH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH/action/storage_attestation","attest_author":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH/action/author_attestation","sign_citation":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH/action/citation_signature","submit_replication":"https://pith.science/pith/ZRXX6SDQAA6WRDMQH7KQJZMAGH/action/replication_record"}},"created_at":"2026-07-05T00:26:11.433897+00:00","updated_at":"2026-07-05T00:26:11.433897+00:00"}