{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:DZHB2UNMJ4ZWC7AFJ2FZY655ZU","short_pith_number":"pith:DZHB2UNM","schema_version":"1.0","canonical_sha256":"1e4e1d51ac4f33617c054e8b9c7bbdcd06687d5fef0ab1ea958f9ce1e9b04a6f","source":{"kind":"arxiv","id":"1911.03872","version":2},"attestation_state":"computed","paper":{"title":"Location Attention for Extrapolation to Longer Sequences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dieuwke Hupkes, Elia Bruni, Gautier Dagan, Yann Dubois","submitted_at":"2019-11-10T07:39:42Z","abstract_excerpt":"Neural networks are surprisingly good at interpolating and perform remarkably well when the training set examples resemble those in the test set. However, they are often unable to extrapolate patterns beyond the seen data, even when the abstractions required for such patterns are simple. In this paper, we first review the notion of extrapolation, why it is important and how one could hope to tackle it. We then focus on a specific type of extrapolation which is especially useful for natural language processing: generalization to sequences that are longer than the training ones. We hypothesize t"},"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":"1911.03872","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-10T07:39:42Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"29206b9182b15fa26cbadd62b202449bf8a27c50fac25859d7e0ebbf392c1d28","abstract_canon_sha256":"8c20475b2d64501f52da68187be67a0b8daa04ece327b88185c39dd40dd8ea44"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:57:24.499484Z","signature_b64":"2ms6hB66QMtnyAlW3aACNooNfx/+MCDdSGweK1f1mgYoI+qpEoFydsK0heOyugWZSycObmEKl/VjGXVxIZJwCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e4e1d51ac4f33617c054e8b9c7bbdcd06687d5fef0ab1ea958f9ce1e9b04a6f","last_reissued_at":"2026-07-05T00:57:24.499118Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:57:24.499118Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Location Attention for Extrapolation to Longer Sequences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dieuwke Hupkes, Elia Bruni, Gautier Dagan, Yann Dubois","submitted_at":"2019-11-10T07:39:42Z","abstract_excerpt":"Neural networks are surprisingly good at interpolating and perform remarkably well when the training set examples resemble those in the test set. However, they are often unable to extrapolate patterns beyond the seen data, even when the abstractions required for such patterns are simple. In this paper, we first review the notion of extrapolation, why it is important and how one could hope to tackle it. We then focus on a specific type of extrapolation which is especially useful for natural language processing: generalization to sequences that are longer than the training ones. We hypothesize t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.03872","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/1911.03872/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":"1911.03872","created_at":"2026-07-05T00:57:24.499175+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.03872v2","created_at":"2026-07-05T00:57:24.499175+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.03872","created_at":"2026-07-05T00:57:24.499175+00:00"},{"alias_kind":"pith_short_12","alias_value":"DZHB2UNMJ4ZW","created_at":"2026-07-05T00:57:24.499175+00:00"},{"alias_kind":"pith_short_16","alias_value":"DZHB2UNMJ4ZWC7AF","created_at":"2026-07-05T00:57:24.499175+00:00"},{"alias_kind":"pith_short_8","alias_value":"DZHB2UNM","created_at":"2026-07-05T00:57:24.499175+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.09251","citing_title":"Extrapolation by Association: Length Generalization Transfer in Transformers","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU","json":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU.json","graph_json":"https://pith.science/api/pith-number/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/graph.json","events_json":"https://pith.science/api/pith-number/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/events.json","paper":"https://pith.science/paper/DZHB2UNM"},"agent_actions":{"view_html":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU","download_json":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU.json","view_paper":"https://pith.science/paper/DZHB2UNM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.03872&json=true","fetch_graph":"https://pith.science/api/pith-number/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/graph.json","fetch_events":"https://pith.science/api/pith-number/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/action/storage_attestation","attest_author":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/action/author_attestation","sign_citation":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/action/citation_signature","submit_replication":"https://pith.science/pith/DZHB2UNMJ4ZWC7AFJ2FZY655ZU/action/replication_record"}},"created_at":"2026-07-05T00:57:24.499175+00:00","updated_at":"2026-07-05T00:57:24.499175+00:00"}