{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:V543UGKCE6KWH57MSBYJKFHLDB","short_pith_number":"pith:V543UGKC","schema_version":"1.0","canonical_sha256":"af79ba1942279563f7ec90709514eb184810389f85a83e1df61ff1859a974d2e","source":{"kind":"arxiv","id":"1910.10387","version":2},"attestation_state":"computed","paper":{"title":"Speech-XLNet: Unsupervised Acoustic Model Pretraining For Self-Attention Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Dan Su, Dong Yu, Guangsen Wang, Helen Meng, Xingchen Song, Yiheng Huang, Zhiyong Wu","submitted_at":"2019-10-23T07:08:03Z","abstract_excerpt":"Self-attention network (SAN) can benefit significantly from the bi-directional representation learning through unsupervised pretraining paradigms such as BERT and XLNet. In this paper, we present an XLNet-like pretraining scheme \"Speech-XLNet\" for unsupervised acoustic model pretraining to learn speech representations with SAN. The pretrained SAN is finetuned under the hybrid SAN/HMM framework. We conjecture that by shuffling the speech frame orders, the permutation in Speech-XLNet serves as a strong regularizer to encourage the SAN to make inferences by focusing on global structures through i"},"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":"1910.10387","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-10-23T07:08:03Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"e6ecb76c5293927ca40211e02791496b2e2ec17e1cfcd5e48f78685f4125579b","abstract_canon_sha256":"90f71343f2b6e178f70573843ff10f4336ba76c6dc7b35f73c4b03452193f8c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:05:00.318988Z","signature_b64":"5U3NF0HwMZiWFxBTsXtg06+vcPSfGE9tw5z7CEx82MZO4Yhqbr5wtsKzBIfcphF3Tjoq5N+8OTVpbYvI4+1nBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af79ba1942279563f7ec90709514eb184810389f85a83e1df61ff1859a974d2e","last_reissued_at":"2026-07-05T01:05:00.318445Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:05:00.318445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Speech-XLNet: Unsupervised Acoustic Model Pretraining For Self-Attention Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Dan Su, Dong Yu, Guangsen Wang, Helen Meng, Xingchen Song, Yiheng Huang, Zhiyong Wu","submitted_at":"2019-10-23T07:08:03Z","abstract_excerpt":"Self-attention network (SAN) can benefit significantly from the bi-directional representation learning through unsupervised pretraining paradigms such as BERT and XLNet. In this paper, we present an XLNet-like pretraining scheme \"Speech-XLNet\" for unsupervised acoustic model pretraining to learn speech representations with SAN. The pretrained SAN is finetuned under the hybrid SAN/HMM framework. We conjecture that by shuffling the speech frame orders, the permutation in Speech-XLNet serves as a strong regularizer to encourage the SAN to make inferences by focusing on global structures through i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.10387","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/1910.10387/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":"1910.10387","created_at":"2026-07-05T01:05:00.318499+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.10387v2","created_at":"2026-07-05T01:05:00.318499+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.10387","created_at":"2026-07-05T01:05:00.318499+00:00"},{"alias_kind":"pith_short_12","alias_value":"V543UGKCE6KW","created_at":"2026-07-05T01:05:00.318499+00:00"},{"alias_kind":"pith_short_16","alias_value":"V543UGKCE6KWH57M","created_at":"2026-07-05T01:05:00.318499+00:00"},{"alias_kind":"pith_short_8","alias_value":"V543UGKC","created_at":"2026-07-05T01:05:00.318499+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/V543UGKCE6KWH57MSBYJKFHLDB","json":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB.json","graph_json":"https://pith.science/api/pith-number/V543UGKCE6KWH57MSBYJKFHLDB/graph.json","events_json":"https://pith.science/api/pith-number/V543UGKCE6KWH57MSBYJKFHLDB/events.json","paper":"https://pith.science/paper/V543UGKC"},"agent_actions":{"view_html":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB","download_json":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB.json","view_paper":"https://pith.science/paper/V543UGKC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.10387&json=true","fetch_graph":"https://pith.science/api/pith-number/V543UGKCE6KWH57MSBYJKFHLDB/graph.json","fetch_events":"https://pith.science/api/pith-number/V543UGKCE6KWH57MSBYJKFHLDB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB/action/storage_attestation","attest_author":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB/action/author_attestation","sign_citation":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB/action/citation_signature","submit_replication":"https://pith.science/pith/V543UGKCE6KWH57MSBYJKFHLDB/action/replication_record"}},"created_at":"2026-07-05T01:05:00.318499+00:00","updated_at":"2026-07-05T01:05:00.318499+00:00"}