{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:O4UE5MLXHPLEMVDH55MNUSPPSX","short_pith_number":"pith:O4UE5MLX","schema_version":"1.0","canonical_sha256":"77284eb1773bd6465467ef58da49ef95cacf6db8d37c0a0074e5b156d50a9c2f","source":{"kind":"arxiv","id":"2210.14431","version":3},"attestation_state":"computed","paper":{"title":"$N$-gram Is Back: Residual Learning of Neural Text Generation with $n$-gram Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deng Cai, Huayang Li, Jin Xu, Taro Watanabe","submitted_at":"2022-10-26T02:42:53Z","abstract_excerpt":"$N$-gram language models (LM) have been largely superseded by neural LMs as the latter exhibits better performance. However, we find that $n$-gram models can achieve satisfactory performance on a large proportion of testing cases, indicating they have already captured abundant knowledge of the language with relatively low computational cost. With this observation, we propose to learn a neural LM that fits the residual between an $n$-gram LM and the real-data distribution. The combination of $n$-gram and neural LMs not only allows the neural part to focus on the deeper understanding of language"},"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":"2210.14431","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-26T02:42:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"54674eeed6a48b7018ace225f4985ad94c7160a426381144970e3dab2e6affba","abstract_canon_sha256":"9077545182495bcd81f32d290f2129febd5a4ca5fb4a48e8bdc3e2693080c1f9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:54.692979Z","signature_b64":"aHzx+eRPexc5bBk8E2XoL5AQHdedf4rnNpCpZMdv2EOcFnruD8hqefCDhCTbhx0ZDHVyfMJL1E73DML90SgEBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77284eb1773bd6465467ef58da49ef95cacf6db8d37c0a0074e5b156d50a9c2f","last_reissued_at":"2026-07-05T05:12:54.692496Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:54.692496Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"$N$-gram Is Back: Residual Learning of Neural Text Generation with $n$-gram Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deng Cai, Huayang Li, Jin Xu, Taro Watanabe","submitted_at":"2022-10-26T02:42:53Z","abstract_excerpt":"$N$-gram language models (LM) have been largely superseded by neural LMs as the latter exhibits better performance. However, we find that $n$-gram models can achieve satisfactory performance on a large proportion of testing cases, indicating they have already captured abundant knowledge of the language with relatively low computational cost. With this observation, we propose to learn a neural LM that fits the residual between an $n$-gram LM and the real-data distribution. The combination of $n$-gram and neural LMs not only allows the neural part to focus on the deeper understanding of language"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.14431","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/2210.14431/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":"2210.14431","created_at":"2026-07-05T05:12:54.692558+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.14431v3","created_at":"2026-07-05T05:12:54.692558+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.14431","created_at":"2026-07-05T05:12:54.692558+00:00"},{"alias_kind":"pith_short_12","alias_value":"O4UE5MLXHPLE","created_at":"2026-07-05T05:12:54.692558+00:00"},{"alias_kind":"pith_short_16","alias_value":"O4UE5MLXHPLEMVDH","created_at":"2026-07-05T05:12:54.692558+00:00"},{"alias_kind":"pith_short_8","alias_value":"O4UE5MLX","created_at":"2026-07-05T05:12:54.692558+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/O4UE5MLXHPLEMVDH55MNUSPPSX","json":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX.json","graph_json":"https://pith.science/api/pith-number/O4UE5MLXHPLEMVDH55MNUSPPSX/graph.json","events_json":"https://pith.science/api/pith-number/O4UE5MLXHPLEMVDH55MNUSPPSX/events.json","paper":"https://pith.science/paper/O4UE5MLX"},"agent_actions":{"view_html":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX","download_json":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX.json","view_paper":"https://pith.science/paper/O4UE5MLX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.14431&json=true","fetch_graph":"https://pith.science/api/pith-number/O4UE5MLXHPLEMVDH55MNUSPPSX/graph.json","fetch_events":"https://pith.science/api/pith-number/O4UE5MLXHPLEMVDH55MNUSPPSX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX/action/storage_attestation","attest_author":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX/action/author_attestation","sign_citation":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX/action/citation_signature","submit_replication":"https://pith.science/pith/O4UE5MLXHPLEMVDH55MNUSPPSX/action/replication_record"}},"created_at":"2026-07-05T05:12:54.692558+00:00","updated_at":"2026-07-05T05:12:54.692558+00:00"}