{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SJG722IUBK64WYYUWT5IMVR72I","short_pith_number":"pith:SJG722IU","schema_version":"1.0","canonical_sha256":"924dfd69140abdcb6314b4fa86563fd235dd6ad383fa97657d7e19afc9e9c65b","source":{"kind":"arxiv","id":"2211.01180","version":2},"attestation_state":"computed","paper":{"title":"M-SpeechCLIP: Leveraging Large-Scale, Pre-Trained Models for Multilingual Speech to Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"David Harwath, Heng-Jui Chang, Hsuan-Fu Wang, Hung-yi Lee, Layne Berry, Yi-Jen Shih","submitted_at":"2022-11-02T14:54:45Z","abstract_excerpt":"This work investigates the use of large-scale, English-only pre-trained models (CLIP and HuBERT) for multilingual image-speech retrieval. For non-English image-speech retrieval, we outperform the current state-of-the-art performance by a wide margin both when training separate models for each language, and with a single model which processes speech in all three languages. We identify key differences in model behavior and performance between English and non-English settings, attributable to the English-only pre-training of CLIP and HuBERT, and investigate how fine-tuning the pre-trained models "},"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":"2211.01180","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-02T14:54:45Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"9e23aba69c55279e4269db0e875f0f02c22a7e7a9b06813205bec67884c9f24e","abstract_canon_sha256":"cc9e2cbb692da95da7cfe1b74dc0e493d37bdfeccb7ab87e362ac686c2aeb9ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:59:21.811424Z","signature_b64":"rtrY5we/UCtlEoZwpy8F7sTlxliLOlOL51oUqh7a762MM3WdIAuFnUhw+lOmrBwWFmMhdIJVhwVsS7VoIOSFCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"924dfd69140abdcb6314b4fa86563fd235dd6ad383fa97657d7e19afc9e9c65b","last_reissued_at":"2026-07-05T05:59:21.811006Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:59:21.811006Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M-SpeechCLIP: Leveraging Large-Scale, Pre-Trained Models for Multilingual Speech to Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"David Harwath, Heng-Jui Chang, Hsuan-Fu Wang, Hung-yi Lee, Layne Berry, Yi-Jen Shih","submitted_at":"2022-11-02T14:54:45Z","abstract_excerpt":"This work investigates the use of large-scale, English-only pre-trained models (CLIP and HuBERT) for multilingual image-speech retrieval. For non-English image-speech retrieval, we outperform the current state-of-the-art performance by a wide margin both when training separate models for each language, and with a single model which processes speech in all three languages. We identify key differences in model behavior and performance between English and non-English settings, attributable to the English-only pre-training of CLIP and HuBERT, and investigate how fine-tuning the pre-trained models "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.01180","kind":"arxiv","version":2},"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/2211.01180/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":"2211.01180","created_at":"2026-07-05T05:59:21.811073+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.01180v2","created_at":"2026-07-05T05:59:21.811073+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.01180","created_at":"2026-07-05T05:59:21.811073+00:00"},{"alias_kind":"pith_short_12","alias_value":"SJG722IUBK64","created_at":"2026-07-05T05:59:21.811073+00:00"},{"alias_kind":"pith_short_16","alias_value":"SJG722IUBK64WYYU","created_at":"2026-07-05T05:59:21.811073+00:00"},{"alias_kind":"pith_short_8","alias_value":"SJG722IU","created_at":"2026-07-05T05:59:21.811073+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/SJG722IUBK64WYYUWT5IMVR72I","json":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I.json","graph_json":"https://pith.science/api/pith-number/SJG722IUBK64WYYUWT5IMVR72I/graph.json","events_json":"https://pith.science/api/pith-number/SJG722IUBK64WYYUWT5IMVR72I/events.json","paper":"https://pith.science/paper/SJG722IU"},"agent_actions":{"view_html":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I","download_json":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I.json","view_paper":"https://pith.science/paper/SJG722IU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.01180&json=true","fetch_graph":"https://pith.science/api/pith-number/SJG722IUBK64WYYUWT5IMVR72I/graph.json","fetch_events":"https://pith.science/api/pith-number/SJG722IUBK64WYYUWT5IMVR72I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I/action/storage_attestation","attest_author":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I/action/author_attestation","sign_citation":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I/action/citation_signature","submit_replication":"https://pith.science/pith/SJG722IUBK64WYYUWT5IMVR72I/action/replication_record"}},"created_at":"2026-07-05T05:59:21.811073+00:00","updated_at":"2026-07-05T05:59:21.811073+00:00"}