{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:6V4RFHPHCLSAUUDWIALJRLR3VD","short_pith_number":"pith:6V4RFHPH","schema_version":"1.0","canonical_sha256":"f579129de712e40a5076401698ae3ba8e0ac53fbcceeb80c5bb0c33f23551e4d","source":{"kind":"arxiv","id":"2102.02557","version":1},"attestation_state":"computed","paper":{"title":"Adaptive Semiparametric Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cyprien de Masson d'Autume, Dani Yogatama, Lingpeng Kong","submitted_at":"2021-02-04T11:47:03Z","abstract_excerpt":"We present a language model that combines a large parametric neural network (i.e., a transformer) with a non-parametric episodic memory component in an integrated architecture. Our model uses extended short-term context by caching local hidden states -- similar to transformer-XL -- and global long-term memory by retrieving a set of nearest neighbor tokens at each timestep. We design a gating function to adaptively combine multiple information sources to make a prediction. This mechanism allows the model to use either local context, short-term memory, or long-term memory (or any combination of "},"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":"2102.02557","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-02-04T11:47:03Z","cross_cats_sorted":[],"title_canon_sha256":"801de53e9f6e31415654e9349b182d2c191ed1893d5ad26468ff1b4571f0adde","abstract_canon_sha256":"a6fa20c9be8f72f060f0f41cd2f8fdbdcfed9808f8e575d28483a4e99a2498dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:12:49.507782Z","signature_b64":"UsLquLAbcNtWeSYNXUAGgmeWHvT97fVAlWQ3tHdf4mWGL124EwqJh3NDJhMtEQ2o3Rypl5MJOrKLN81gJSItDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f579129de712e40a5076401698ae3ba8e0ac53fbcceeb80c5bb0c33f23551e4d","last_reissued_at":"2026-07-05T02:12:49.507332Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:12:49.507332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive Semiparametric Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cyprien de Masson d'Autume, Dani Yogatama, Lingpeng Kong","submitted_at":"2021-02-04T11:47:03Z","abstract_excerpt":"We present a language model that combines a large parametric neural network (i.e., a transformer) with a non-parametric episodic memory component in an integrated architecture. Our model uses extended short-term context by caching local hidden states -- similar to transformer-XL -- and global long-term memory by retrieving a set of nearest neighbor tokens at each timestep. We design a gating function to adaptively combine multiple information sources to make a prediction. This mechanism allows the model to use either local context, short-term memory, or long-term memory (or any combination of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.02557","kind":"arxiv","version":1},"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/2102.02557/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":"2102.02557","created_at":"2026-07-05T02:12:49.507387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.02557v1","created_at":"2026-07-05T02:12:49.507387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.02557","created_at":"2026-07-05T02:12:49.507387+00:00"},{"alias_kind":"pith_short_12","alias_value":"6V4RFHPHCLSA","created_at":"2026-07-05T02:12:49.507387+00:00"},{"alias_kind":"pith_short_16","alias_value":"6V4RFHPHCLSAUUDW","created_at":"2026-07-05T02:12:49.507387+00:00"},{"alias_kind":"pith_short_8","alias_value":"6V4RFHPH","created_at":"2026-07-05T02:12:49.507387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.02303","citing_title":"A Hippocampus for Linear Attention: An Exact Memory for What the Recurrent State Forgets","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD","json":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD.json","graph_json":"https://pith.science/api/pith-number/6V4RFHPHCLSAUUDWIALJRLR3VD/graph.json","events_json":"https://pith.science/api/pith-number/6V4RFHPHCLSAUUDWIALJRLR3VD/events.json","paper":"https://pith.science/paper/6V4RFHPH"},"agent_actions":{"view_html":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD","download_json":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD.json","view_paper":"https://pith.science/paper/6V4RFHPH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.02557&json=true","fetch_graph":"https://pith.science/api/pith-number/6V4RFHPHCLSAUUDWIALJRLR3VD/graph.json","fetch_events":"https://pith.science/api/pith-number/6V4RFHPHCLSAUUDWIALJRLR3VD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD/action/storage_attestation","attest_author":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD/action/author_attestation","sign_citation":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD/action/citation_signature","submit_replication":"https://pith.science/pith/6V4RFHPHCLSAUUDWIALJRLR3VD/action/replication_record"}},"created_at":"2026-07-05T02:12:49.507387+00:00","updated_at":"2026-07-05T02:12:49.507387+00:00"}