{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ZL5J2VQZMOVIZZ4IGLW5NWYKZZ","short_pith_number":"pith:ZL5J2VQZ","schema_version":"1.0","canonical_sha256":"cafa9d561963aa8ce78832edd6db0ace67c93ec6ae84b205ca8708873be770d4","source":{"kind":"arxiv","id":"2102.04754","version":1},"attestation_state":"computed","paper":{"title":"Bayesian Transformer Language Models for Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boyang Xue, Helen Meng, Jianwei Yu, Junhao Xu, Mengzhe Geng, Shansong Liu, Shoukang Hu, Xunying Liu, Zi Ye","submitted_at":"2021-02-09T10:55:27Z","abstract_excerpt":"State-of-the-art neural language models (LMs) represented by Transformers are highly complex. Their use of fixed, deterministic parameter estimates fail to account for model uncertainty and lead to over-fitting and poor generalization when given limited training data. In order to address these issues, this paper proposes a full Bayesian learning framework for Transformer LM estimation. Efficient variational inference based approaches are used to estimate the latent parameter posterior distributions associated with different parts of the Transformer model architecture including multi-head self-"},"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":"2102.04754","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-09T10:55:27Z","cross_cats_sorted":[],"title_canon_sha256":"c03997691b55cb74d6bc8ef699214c94cb083fb8cf5adb6cd62887d855ea192f","abstract_canon_sha256":"935ac60e91b6398aead939f92cbf86289f0ba37c50c4c34a64756e865a872104"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:14:04.409201Z","signature_b64":"CiTXVtAbrty4IATOEfGkfB30RqUatg7AnM0wbpZ6tb9+reGDTct6udfOE/rWJDaTd1P89LE+YG+d7Gp7WdOVDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cafa9d561963aa8ce78832edd6db0ace67c93ec6ae84b205ca8708873be770d4","last_reissued_at":"2026-07-05T02:14:04.408749Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:14:04.408749Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian Transformer Language Models for Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boyang Xue, Helen Meng, Jianwei Yu, Junhao Xu, Mengzhe Geng, Shansong Liu, Shoukang Hu, Xunying Liu, Zi Ye","submitted_at":"2021-02-09T10:55:27Z","abstract_excerpt":"State-of-the-art neural language models (LMs) represented by Transformers are highly complex. Their use of fixed, deterministic parameter estimates fail to account for model uncertainty and lead to over-fitting and poor generalization when given limited training data. In order to address these issues, this paper proposes a full Bayesian learning framework for Transformer LM estimation. Efficient variational inference based approaches are used to estimate the latent parameter posterior distributions associated with different parts of the Transformer model architecture including multi-head self-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.04754","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/2102.04754/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":"2102.04754","created_at":"2026-07-05T02:14:04.408807+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.04754v1","created_at":"2026-07-05T02:14:04.408807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.04754","created_at":"2026-07-05T02:14:04.408807+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZL5J2VQZMOVI","created_at":"2026-07-05T02:14:04.408807+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZL5J2VQZMOVIZZ4I","created_at":"2026-07-05T02:14:04.408807+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZL5J2VQZ","created_at":"2026-07-05T02:14:04.408807+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/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ","json":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ.json","graph_json":"https://pith.science/api/pith-number/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/graph.json","events_json":"https://pith.science/api/pith-number/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/events.json","paper":"https://pith.science/paper/ZL5J2VQZ"},"agent_actions":{"view_html":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ","download_json":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ.json","view_paper":"https://pith.science/paper/ZL5J2VQZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.04754&json=true","fetch_graph":"https://pith.science/api/pith-number/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/action/storage_attestation","attest_author":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/action/author_attestation","sign_citation":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/action/citation_signature","submit_replication":"https://pith.science/pith/ZL5J2VQZMOVIZZ4IGLW5NWYKZZ/action/replication_record"}},"created_at":"2026-07-05T02:14:04.408807+00:00","updated_at":"2026-07-05T02:14:04.408807+00:00"}