{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:J5UUUCOPLLUYGL2UWRRAEKB2GL","short_pith_number":"pith:J5UUUCOP","schema_version":"1.0","canonical_sha256":"4f694a09cf5ae9832f54b46202283a32e66512ed090216ec214bce3f3b00c5d5","source":{"kind":"arxiv","id":"2409.11274","version":3},"attestation_state":"computed","paper":{"title":"Task Arithmetic for Language Expansion in Speech Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Emiru Tsunoo, Hayato Futami, Shinji Watanabe, Siddhant Arora, Wen Shen Teo, Yao-Fei Cheng, Yosuke Kashiwagi","submitted_at":"2024-09-17T15:25:11Z","abstract_excerpt":"Recent progress in large language models (LLMs) has gained interest in speech-text multimodal foundation models, achieving strong performance on instruction-tuned speech translation (ST). However, expanding language pairs is costly due to re-training on combined new and previous datasets. To address this, we aim to build a one-to-many ST system from existing one-to-one ST systems using task arithmetic without re-training. Direct application of task arithmetic in ST leads to language confusion; therefore, we introduce an augmented task arithmetic method incorporating a language control model to"},"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":"2409.11274","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-17T15:25:11Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"765e3a93dc6bdef78a0016dfdfa313c9f8b92ce60d61dd2cf50b6ae8f7ef70d7","abstract_canon_sha256":"d5be1109d08b30559dfca9ced3edbecfd3247f1b3d634988746fdf7cf535edba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:40.862767Z","signature_b64":"r0O+5b6fhET8Na7MUe7w3BaIOuzEtCb7mHT7i3LO1z8wjeFaNnK7N/lBk6NLKaYmW0uAyZr4MrmX2ohiG7owAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f694a09cf5ae9832f54b46202283a32e66512ed090216ec214bce3f3b00c5d5","last_reissued_at":"2026-07-05T11:44:40.862335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:40.862335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Task Arithmetic for Language Expansion in Speech Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Emiru Tsunoo, Hayato Futami, Shinji Watanabe, Siddhant Arora, Wen Shen Teo, Yao-Fei Cheng, Yosuke Kashiwagi","submitted_at":"2024-09-17T15:25:11Z","abstract_excerpt":"Recent progress in large language models (LLMs) has gained interest in speech-text multimodal foundation models, achieving strong performance on instruction-tuned speech translation (ST). However, expanding language pairs is costly due to re-training on combined new and previous datasets. To address this, we aim to build a one-to-many ST system from existing one-to-one ST systems using task arithmetic without re-training. Direct application of task arithmetic in ST leads to language confusion; therefore, we introduce an augmented task arithmetic method incorporating a language control model to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.11274","kind":"arxiv","version":3},"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/2409.11274/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":"2409.11274","created_at":"2026-07-05T11:44:40.862397+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.11274v3","created_at":"2026-07-05T11:44:40.862397+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.11274","created_at":"2026-07-05T11:44:40.862397+00:00"},{"alias_kind":"pith_short_12","alias_value":"J5UUUCOPLLUY","created_at":"2026-07-05T11:44:40.862397+00:00"},{"alias_kind":"pith_short_16","alias_value":"J5UUUCOPLLUYGL2U","created_at":"2026-07-05T11:44:40.862397+00:00"},{"alias_kind":"pith_short_8","alias_value":"J5UUUCOP","created_at":"2026-07-05T11:44:40.862397+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.12672","citing_title":"Speech-FT: Merging Pre-trained And Fine-Tuned Speech Representation Models For Cross-Task Generalization","ref_index":57,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL","json":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL.json","graph_json":"https://pith.science/api/pith-number/J5UUUCOPLLUYGL2UWRRAEKB2GL/graph.json","events_json":"https://pith.science/api/pith-number/J5UUUCOPLLUYGL2UWRRAEKB2GL/events.json","paper":"https://pith.science/paper/J5UUUCOP"},"agent_actions":{"view_html":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL","download_json":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL.json","view_paper":"https://pith.science/paper/J5UUUCOP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.11274&json=true","fetch_graph":"https://pith.science/api/pith-number/J5UUUCOPLLUYGL2UWRRAEKB2GL/graph.json","fetch_events":"https://pith.science/api/pith-number/J5UUUCOPLLUYGL2UWRRAEKB2GL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL/action/storage_attestation","attest_author":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL/action/author_attestation","sign_citation":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL/action/citation_signature","submit_replication":"https://pith.science/pith/J5UUUCOPLLUYGL2UWRRAEKB2GL/action/replication_record"}},"created_at":"2026-07-05T11:44:40.862397+00:00","updated_at":"2026-07-05T11:44:40.862397+00:00"}