{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GLS7R4AHYHZSQPMFLR6CEPGWY7","short_pith_number":"pith:GLS7R4AH","schema_version":"1.0","canonical_sha256":"32e5f8f007c1f3283d855c7c223cd6c7c2ece7f99bd013d314f9c559c5f20f09","source":{"kind":"arxiv","id":"2509.09701","version":1},"attestation_state":"computed","paper":{"title":"Optimal Multi-Task Learning at Regularization Horizon for Speech Translation Task","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jungho Jung, Junhyun Lee","submitted_at":"2025-09-04T17:21:36Z","abstract_excerpt":"End-to-end speech-to-text translation typically suffers from the scarcity of paired speech-text data. One way to overcome this shortcoming is to utilize the bitext data from the Machine Translation (MT) task and perform Multi-Task Learning (MTL). In this paper, we formulate MTL from a regularization perspective and explore how sequences can be regularized within and across modalities. By thoroughly investigating the effect of consistency regularization (different modality) and R-drop (same modality), we show how they respectively contribute to the total regularization. We also demonstrate that"},"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":"2509.09701","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T17:21:36Z","cross_cats_sorted":[],"title_canon_sha256":"8bff0f33cc99a76b80d398ca076ff8be785c3b013d761d5289c5f6f3a89f34cb","abstract_canon_sha256":"4b5986c6aa441896833202d9fa5089542447843493aaa99aefaf66d4b9999709"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:47.407289Z","signature_b64":"fuKKmDbP2J4hL/VIbTw/+S2BunesTLxMmIebvcd5yGu99Mrt0Of4Z0Qprzv8yHHBt2J6rZx9N0IIxRrH8qOaCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32e5f8f007c1f3283d855c7c223cd6c7c2ece7f99bd013d314f9c559c5f20f09","last_reissued_at":"2026-07-05T12:09:47.406817Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:47.406817Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimal Multi-Task Learning at Regularization Horizon for Speech Translation Task","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jungho Jung, Junhyun Lee","submitted_at":"2025-09-04T17:21:36Z","abstract_excerpt":"End-to-end speech-to-text translation typically suffers from the scarcity of paired speech-text data. One way to overcome this shortcoming is to utilize the bitext data from the Machine Translation (MT) task and perform Multi-Task Learning (MTL). In this paper, we formulate MTL from a regularization perspective and explore how sequences can be regularized within and across modalities. By thoroughly investigating the effect of consistency regularization (different modality) and R-drop (same modality), we show how they respectively contribute to the total regularization. We also demonstrate that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.09701","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/2509.09701/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":"2509.09701","created_at":"2026-07-05T12:09:47.406875+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.09701v1","created_at":"2026-07-05T12:09:47.406875+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.09701","created_at":"2026-07-05T12:09:47.406875+00:00"},{"alias_kind":"pith_short_12","alias_value":"GLS7R4AHYHZS","created_at":"2026-07-05T12:09:47.406875+00:00"},{"alias_kind":"pith_short_16","alias_value":"GLS7R4AHYHZSQPMF","created_at":"2026-07-05T12:09:47.406875+00:00"},{"alias_kind":"pith_short_8","alias_value":"GLS7R4AH","created_at":"2026-07-05T12:09:47.406875+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/GLS7R4AHYHZSQPMFLR6CEPGWY7","json":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7.json","graph_json":"https://pith.science/api/pith-number/GLS7R4AHYHZSQPMFLR6CEPGWY7/graph.json","events_json":"https://pith.science/api/pith-number/GLS7R4AHYHZSQPMFLR6CEPGWY7/events.json","paper":"https://pith.science/paper/GLS7R4AH"},"agent_actions":{"view_html":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7","download_json":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7.json","view_paper":"https://pith.science/paper/GLS7R4AH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.09701&json=true","fetch_graph":"https://pith.science/api/pith-number/GLS7R4AHYHZSQPMFLR6CEPGWY7/graph.json","fetch_events":"https://pith.science/api/pith-number/GLS7R4AHYHZSQPMFLR6CEPGWY7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7/action/storage_attestation","attest_author":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7/action/author_attestation","sign_citation":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7/action/citation_signature","submit_replication":"https://pith.science/pith/GLS7R4AHYHZSQPMFLR6CEPGWY7/action/replication_record"}},"created_at":"2026-07-05T12:09:47.406875+00:00","updated_at":"2026-07-05T12:09:47.406875+00:00"}