{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6VBI2BZ47ZFBCPDHMBZHHEMUT5","short_pith_number":"pith:6VBI2BZ4","schema_version":"1.0","canonical_sha256":"f5428d073cfe4a113c6760727391949f4b640ad71f01897e7bb0389922f0f30f","source":{"kind":"arxiv","id":"2509.09196","version":1},"attestation_state":"computed","paper":{"title":"Efficient Trie-based Biasing using K-step Prediction for Rare Word Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chin Yuen Kwok, Jia Qi Yip","submitted_at":"2025-09-11T07:11:46Z","abstract_excerpt":"Contextual biasing improves rare word recognition of ASR models by prioritizing the output of rare words during decoding. A common approach is Trie-based biasing, which gives \"bonus scores\" to partial hypothesis (e.g. \"Bon\") that may lead to the generation of the rare word (e.g. \"Bonham\"). If the full word (\"Bonham\") isn't ultimately recognized, the system revokes those earlier bonuses. This revocation is limited to beam search and is computationally expensive, particularly for models with large decoders. To overcome these limitations, we propose adapting ASR models to look ahead and predict m"},"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":"2509.09196","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-11T07:11:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1aa18e6d1672ccc49b3897072cef4fc7108728cda4374494c91a07ca17f14e89","abstract_canon_sha256":"b3a5ab5939896113450f7771eaefbbfc696d72075b97203f862ecbd1a0d65424"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:31.432712Z","signature_b64":"uUk+VHxhaYmqNZsKCaJ5Ft+7N11sLVsVpoZSDbdma2BZxnV0RApMAam8LArWFxMTrX5RydVg/jMja1KzvMixDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5428d073cfe4a113c6760727391949f4b640ad71f01897e7bb0389922f0f30f","last_reissued_at":"2026-07-05T12:09:31.432136Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:31.432136Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Trie-based Biasing using K-step Prediction for Rare Word Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chin Yuen Kwok, Jia Qi Yip","submitted_at":"2025-09-11T07:11:46Z","abstract_excerpt":"Contextual biasing improves rare word recognition of ASR models by prioritizing the output of rare words during decoding. A common approach is Trie-based biasing, which gives \"bonus scores\" to partial hypothesis (e.g. \"Bon\") that may lead to the generation of the rare word (e.g. \"Bonham\"). If the full word (\"Bonham\") isn't ultimately recognized, the system revokes those earlier bonuses. This revocation is limited to beam search and is computationally expensive, particularly for models with large decoders. To overcome these limitations, we propose adapting ASR models to look ahead and predict m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.09196","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/2509.09196/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":"2509.09196","created_at":"2026-07-05T12:09:31.432204+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.09196v1","created_at":"2026-07-05T12:09:31.432204+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.09196","created_at":"2026-07-05T12:09:31.432204+00:00"},{"alias_kind":"pith_short_12","alias_value":"6VBI2BZ47ZFB","created_at":"2026-07-05T12:09:31.432204+00:00"},{"alias_kind":"pith_short_16","alias_value":"6VBI2BZ47ZFBCPDH","created_at":"2026-07-05T12:09:31.432204+00:00"},{"alias_kind":"pith_short_8","alias_value":"6VBI2BZ4","created_at":"2026-07-05T12:09:31.432204+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/6VBI2BZ47ZFBCPDHMBZHHEMUT5","json":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5.json","graph_json":"https://pith.science/api/pith-number/6VBI2BZ47ZFBCPDHMBZHHEMUT5/graph.json","events_json":"https://pith.science/api/pith-number/6VBI2BZ47ZFBCPDHMBZHHEMUT5/events.json","paper":"https://pith.science/paper/6VBI2BZ4"},"agent_actions":{"view_html":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5","download_json":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5.json","view_paper":"https://pith.science/paper/6VBI2BZ4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.09196&json=true","fetch_graph":"https://pith.science/api/pith-number/6VBI2BZ47ZFBCPDHMBZHHEMUT5/graph.json","fetch_events":"https://pith.science/api/pith-number/6VBI2BZ47ZFBCPDHMBZHHEMUT5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5/action/storage_attestation","attest_author":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5/action/author_attestation","sign_citation":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5/action/citation_signature","submit_replication":"https://pith.science/pith/6VBI2BZ47ZFBCPDHMBZHHEMUT5/action/replication_record"}},"created_at":"2026-07-05T12:09:31.432204+00:00","updated_at":"2026-07-05T12:09:31.432204+00:00"}