{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WW3Z5P7ICIC6JUJ6RLHPFVKIIV","short_pith_number":"pith:WW3Z5P7I","schema_version":"1.0","canonical_sha256":"b5b79ebfe81205e4d13e8acef2d548456ef2ef25973b46ba0298f980ab823df8","source":{"kind":"arxiv","id":"2306.15815","version":1},"attestation_state":"computed","paper":{"title":"Scaling Laws for Discriminative Speech Recognition Rescoring Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Ankur Gandhe, Ariya Rastrow, Ivan Bulyko, Jari Kolehmainen, Prashanth Gurunath Shivakumar, Yile Gu","submitted_at":"2023-06-27T22:19:44Z","abstract_excerpt":"Recent studies have found that model performance has a smooth power-law relationship, or scaling laws, with training data and model size, for a wide range of problems. These scaling laws allow one to choose nearly optimal data and model sizes. We study whether this scaling property is also applicable to second-pass rescoring, which is an important component of speech recognition systems. We focus on RescoreBERT as the rescoring model, which uses a pre-trained Transformer-based architecture fined tuned with an ASR discriminative loss. Using such a rescoring model, we show that the word error ra"},"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":"2306.15815","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2023-06-27T22:19:44Z","cross_cats_sorted":[],"title_canon_sha256":"faf58b8c8af359e37a674157fcfdc7d78a2e14243cc83801ff57096a847b50a0","abstract_canon_sha256":"2b00f545d70b0a469b7f87f9492325426ea1f2c941578f3f6a33e58e81498a14"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:25:44.697033Z","signature_b64":"cX309zJj0ZLS4adb1x/EpBi2ChAfHuUMG7ZJFGUy6tjROka+0GIcUWUvebfRPw9yCUvCVqA7Y6lnJGCn7WXPDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b5b79ebfe81205e4d13e8acef2d548456ef2ef25973b46ba0298f980ab823df8","last_reissued_at":"2026-07-05T06:25:44.696459Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:25:44.696459Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Laws for Discriminative Speech Recognition Rescoring Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Ankur Gandhe, Ariya Rastrow, Ivan Bulyko, Jari Kolehmainen, Prashanth Gurunath Shivakumar, Yile Gu","submitted_at":"2023-06-27T22:19:44Z","abstract_excerpt":"Recent studies have found that model performance has a smooth power-law relationship, or scaling laws, with training data and model size, for a wide range of problems. These scaling laws allow one to choose nearly optimal data and model sizes. We study whether this scaling property is also applicable to second-pass rescoring, which is an important component of speech recognition systems. We focus on RescoreBERT as the rescoring model, which uses a pre-trained Transformer-based architecture fined tuned with an ASR discriminative loss. Using such a rescoring model, we show that the word error ra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.15815","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/2306.15815/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":"2306.15815","created_at":"2026-07-05T06:25:44.696520+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.15815v1","created_at":"2026-07-05T06:25:44.696520+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.15815","created_at":"2026-07-05T06:25:44.696520+00:00"},{"alias_kind":"pith_short_12","alias_value":"WW3Z5P7ICIC6","created_at":"2026-07-05T06:25:44.696520+00:00"},{"alias_kind":"pith_short_16","alias_value":"WW3Z5P7ICIC6JUJ6","created_at":"2026-07-05T06:25:44.696520+00:00"},{"alias_kind":"pith_short_8","alias_value":"WW3Z5P7I","created_at":"2026-07-05T06:25:44.696520+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/WW3Z5P7ICIC6JUJ6RLHPFVKIIV","json":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV.json","graph_json":"https://pith.science/api/pith-number/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/graph.json","events_json":"https://pith.science/api/pith-number/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/events.json","paper":"https://pith.science/paper/WW3Z5P7I"},"agent_actions":{"view_html":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV","download_json":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV.json","view_paper":"https://pith.science/paper/WW3Z5P7I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.15815&json=true","fetch_graph":"https://pith.science/api/pith-number/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/graph.json","fetch_events":"https://pith.science/api/pith-number/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/action/storage_attestation","attest_author":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/action/author_attestation","sign_citation":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/action/citation_signature","submit_replication":"https://pith.science/pith/WW3Z5P7ICIC6JUJ6RLHPFVKIIV/action/replication_record"}},"created_at":"2026-07-05T06:25:44.696520+00:00","updated_at":"2026-07-05T06:25:44.696520+00:00"}