{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2014:NSNKVNYVSLPKX46RGZSI43MB4Q","short_pith_number":"pith:NSNKVNYV","schema_version":"1.0","canonical_sha256":"6c9aaab71592deabf3d136648e6d81e43f9b6ee98094822cea7690bb84b09d53","source":{"kind":"arxiv","id":"1412.1454","version":2},"attestation_state":"computed","paper":{"title":"Skip-gram Language Modeling Using Sparse Non-negative Matrix Probability Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Ciprian Chelba, Joris Pelemans, Noam Shazeer","submitted_at":"2014-12-03T19:42:12Z","abstract_excerpt":"We present a novel family of language model (LM) estimation techniques named Sparse Non-negative Matrix (SNM) estimation. A first set of experiments empirically evaluating it on the One Billion Word Benchmark shows that SNM $n$-gram LMs perform almost as well as the well-established Kneser-Ney (KN) models. When using skip-gram features the models are able to match the state-of-the-art recurrent neural network (RNN) LMs; combining the two modeling techniques yields the best known result on the benchmark. The computational advantages of SNM over both maximum entropy and RNN LM estimation are pro"},"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":"1412.1454","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2014-12-03T19:42:12Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"d6a32b54d6f93db1944e9ffa988cbcc50b5b0b251e3f495e745a92478be3030a","abstract_canon_sha256":"d896b026dff5dc7b737f58e5bd18b6a377381ffb6fb7e5c17ccd474e86c76d0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:37:47.377569Z","signature_b64":"31stw9x0L9MsYZCgtlDK6YvlLl0eLcUlHIGcYZaCqXzKUebH48ywbfxZuQLveXN2ojsvGN3EbQ/Q0CRAg07oCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c9aaab71592deabf3d136648e6d81e43f9b6ee98094822cea7690bb84b09d53","last_reissued_at":"2026-05-18T01:37:47.377076Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:37:47.377076Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Skip-gram Language Modeling Using Sparse Non-negative Matrix Probability Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Ciprian Chelba, Joris Pelemans, Noam Shazeer","submitted_at":"2014-12-03T19:42:12Z","abstract_excerpt":"We present a novel family of language model (LM) estimation techniques named Sparse Non-negative Matrix (SNM) estimation. A first set of experiments empirically evaluating it on the One Billion Word Benchmark shows that SNM $n$-gram LMs perform almost as well as the well-established Kneser-Ney (KN) models. When using skip-gram features the models are able to match the state-of-the-art recurrent neural network (RNN) LMs; combining the two modeling techniques yields the best known result on the benchmark. The computational advantages of SNM over both maximum entropy and RNN LM estimation are pro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1412.1454","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":""},"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":"1412.1454","created_at":"2026-05-18T01:37:47.377147+00:00"},{"alias_kind":"arxiv_version","alias_value":"1412.1454v2","created_at":"2026-05-18T01:37:47.377147+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1412.1454","created_at":"2026-05-18T01:37:47.377147+00:00"},{"alias_kind":"pith_short_12","alias_value":"NSNKVNYVSLPK","created_at":"2026-05-18T12:28:41.024544+00:00"},{"alias_kind":"pith_short_16","alias_value":"NSNKVNYVSLPKX46R","created_at":"2026-05-18T12:28:41.024544+00:00"},{"alias_kind":"pith_short_8","alias_value":"NSNKVNYV","created_at":"2026-05-18T12:28:41.024544+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/NSNKVNYVSLPKX46RGZSI43MB4Q","json":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q.json","graph_json":"https://pith.science/api/pith-number/NSNKVNYVSLPKX46RGZSI43MB4Q/graph.json","events_json":"https://pith.science/api/pith-number/NSNKVNYVSLPKX46RGZSI43MB4Q/events.json","paper":"https://pith.science/paper/NSNKVNYV"},"agent_actions":{"view_html":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q","download_json":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q.json","view_paper":"https://pith.science/paper/NSNKVNYV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1412.1454&json=true","fetch_graph":"https://pith.science/api/pith-number/NSNKVNYVSLPKX46RGZSI43MB4Q/graph.json","fetch_events":"https://pith.science/api/pith-number/NSNKVNYVSLPKX46RGZSI43MB4Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q/action/storage_attestation","attest_author":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q/action/author_attestation","sign_citation":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q/action/citation_signature","submit_replication":"https://pith.science/pith/NSNKVNYVSLPKX46RGZSI43MB4Q/action/replication_record"}},"created_at":"2026-05-18T01:37:47.377147+00:00","updated_at":"2026-05-18T01:37:47.377147+00:00"}