{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NH7D2GE3PRKZKCSV53QHO25WBO","short_pith_number":"pith:NH7D2GE3","schema_version":"1.0","canonical_sha256":"69fe3d189b7c55950a55eee0776bb60b8cc6f4c70075233c1b7af4b462ee55eb","source":{"kind":"arxiv","id":"2212.03657","version":1},"attestation_state":"computed","paper":{"title":"M3ST: Mix at Three Levels for Speech Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Fengpeng Yue, Mingxuan Wang, Qianqian Dong, Tom Ko, Xuxin Cheng, Yuexian Zou","submitted_at":"2022-12-07T14:22:00Z","abstract_excerpt":"How to solve the data scarcity problem for end-to-end speech-to-text translation (ST)? It's well known that data augmentation is an efficient method to improve performance for many tasks by enlarging the dataset. In this paper, we propose Mix at three levels for Speech Translation (M^3ST) method to increase the diversity of the augmented training corpus. Specifically, we conduct two phases of fine-tuning based on a pre-trained model using external machine translation (MT) data. In the first stage of fine-tuning, we mix the training corpus at three levels, including word level, sentence level a"},"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":"2212.03657","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-07T14:22:00Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"a07cef4abff7a99fcb0bcf1db59621d67c2cc32c971d9d18dcf92c5cfdbd3a25","abstract_canon_sha256":"bb517a52656da3bebc8c3681ce221fb223e857aecd96d91d8d9e9cfa9aa97f40"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:23:19.182902Z","signature_b64":"hR4XnR5fEjwzVawMcLAdJiHRwiLVC47+9WN35uqEgqr/lXu9Y0o4tBSOk1taO2+fB8PG9Gycwh3mMwHgfyXVBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69fe3d189b7c55950a55eee0776bb60b8cc6f4c70075233c1b7af4b462ee55eb","last_reissued_at":"2026-07-05T05:23:19.182471Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:23:19.182471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M3ST: Mix at Three Levels for Speech Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Fengpeng Yue, Mingxuan Wang, Qianqian Dong, Tom Ko, Xuxin Cheng, Yuexian Zou","submitted_at":"2022-12-07T14:22:00Z","abstract_excerpt":"How to solve the data scarcity problem for end-to-end speech-to-text translation (ST)? It's well known that data augmentation is an efficient method to improve performance for many tasks by enlarging the dataset. In this paper, we propose Mix at three levels for Speech Translation (M^3ST) method to increase the diversity of the augmented training corpus. Specifically, we conduct two phases of fine-tuning based on a pre-trained model using external machine translation (MT) data. In the first stage of fine-tuning, we mix the training corpus at three levels, including word level, sentence level a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.03657","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/2212.03657/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":"2212.03657","created_at":"2026-07-05T05:23:19.182535+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.03657v1","created_at":"2026-07-05T05:23:19.182535+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.03657","created_at":"2026-07-05T05:23:19.182535+00:00"},{"alias_kind":"pith_short_12","alias_value":"NH7D2GE3PRKZ","created_at":"2026-07-05T05:23:19.182535+00:00"},{"alias_kind":"pith_short_16","alias_value":"NH7D2GE3PRKZKCSV","created_at":"2026-07-05T05:23:19.182535+00:00"},{"alias_kind":"pith_short_8","alias_value":"NH7D2GE3","created_at":"2026-07-05T05:23:19.182535+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/NH7D2GE3PRKZKCSV53QHO25WBO","json":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO.json","graph_json":"https://pith.science/api/pith-number/NH7D2GE3PRKZKCSV53QHO25WBO/graph.json","events_json":"https://pith.science/api/pith-number/NH7D2GE3PRKZKCSV53QHO25WBO/events.json","paper":"https://pith.science/paper/NH7D2GE3"},"agent_actions":{"view_html":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO","download_json":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO.json","view_paper":"https://pith.science/paper/NH7D2GE3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.03657&json=true","fetch_graph":"https://pith.science/api/pith-number/NH7D2GE3PRKZKCSV53QHO25WBO/graph.json","fetch_events":"https://pith.science/api/pith-number/NH7D2GE3PRKZKCSV53QHO25WBO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO/action/storage_attestation","attest_author":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO/action/author_attestation","sign_citation":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO/action/citation_signature","submit_replication":"https://pith.science/pith/NH7D2GE3PRKZKCSV53QHO25WBO/action/replication_record"}},"created_at":"2026-07-05T05:23:19.182535+00:00","updated_at":"2026-07-05T05:23:19.182535+00:00"}