{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:RC2VCJ5PQGRMTU26BQJUDDUTCQ","short_pith_number":"pith:RC2VCJ5P","schema_version":"1.0","canonical_sha256":"88b55127af81a2c9d35e0c13418e931430f01d4e4d3fd468cb4d6ebefc8ff2c1","source":{"kind":"arxiv","id":"2203.10752","version":3},"attestation_state":"computed","paper":{"title":"XTREME-S: Evaluating Cross-lingual Speech Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexis Conneau, Ankur Bapna, Anton Lozhkov, Clara Rivera, Colin Cherry, Daan van Esch, Jason Riesa, Jonathan H. Clark, Melvin Johnson, Michael Auli, Mihir Kale, Min Ma, Orhan Firat, Patrick von Platen, Sebastian Ruder, Simran Khanuja, Vera Axelrod, Ye Jia, Yu Zhang","submitted_at":"2022-03-21T06:50:21Z","abstract_excerpt":"We introduce XTREME-S, a new benchmark to evaluate universal cross-lingual speech representations in many languages. XTREME-S covers four task families: speech recognition, classification, speech-to-text translation and retrieval. Covering 102 languages from 10+ language families, 3 different domains and 4 task families, XTREME-S aims to simplify multilingual speech representation evaluation, as well as catalyze research in \"universal\" speech representation learning. This paper describes the new benchmark and establishes the first speech-only and speech-text baselines using XLS-R and mSLAM on "},"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":"2203.10752","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-21T06:50:21Z","cross_cats_sorted":[],"title_canon_sha256":"51b0c7b47262ff4f191c9bb7cca22b02ceb876c3912411383d27d516d44bb42c","abstract_canon_sha256":"96f1033bdff68d8a46a5cdf03c76d4f3099bdb152d826865cdb464ad7f46ac54"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:14:28.134784Z","signature_b64":"rrNRM+ME+DTFGntqVTdYHia3zVodS48jtTqFKppaClpjOWA6WaEsvQ7Q2N8DvKko3reAzBjl8H6pTXfec6wWCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88b55127af81a2c9d35e0c13418e931430f01d4e4d3fd468cb4d6ebefc8ff2c1","last_reissued_at":"2026-07-05T04:14:28.134420Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:14:28.134420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"XTREME-S: Evaluating Cross-lingual Speech Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexis Conneau, Ankur Bapna, Anton Lozhkov, Clara Rivera, Colin Cherry, Daan van Esch, Jason Riesa, Jonathan H. Clark, Melvin Johnson, Michael Auli, Mihir Kale, Min Ma, Orhan Firat, Patrick von Platen, Sebastian Ruder, Simran Khanuja, Vera Axelrod, Ye Jia, Yu Zhang","submitted_at":"2022-03-21T06:50:21Z","abstract_excerpt":"We introduce XTREME-S, a new benchmark to evaluate universal cross-lingual speech representations in many languages. XTREME-S covers four task families: speech recognition, classification, speech-to-text translation and retrieval. Covering 102 languages from 10+ language families, 3 different domains and 4 task families, XTREME-S aims to simplify multilingual speech representation evaluation, as well as catalyze research in \"universal\" speech representation learning. This paper describes the new benchmark and establishes the first speech-only and speech-text baselines using XLS-R and mSLAM on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.10752","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/2203.10752/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":"2203.10752","created_at":"2026-07-05T04:14:28.134473+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.10752v3","created_at":"2026-07-05T04:14:28.134473+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.10752","created_at":"2026-07-05T04:14:28.134473+00:00"},{"alias_kind":"pith_short_12","alias_value":"RC2VCJ5PQGRM","created_at":"2026-07-05T04:14:28.134473+00:00"},{"alias_kind":"pith_short_16","alias_value":"RC2VCJ5PQGRMTU26","created_at":"2026-07-05T04:14:28.134473+00:00"},{"alias_kind":"pith_short_8","alias_value":"RC2VCJ5P","created_at":"2026-07-05T04:14:28.134473+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08194","citing_title":"GlobeAudio: A Multilingual Multicultural Benchmark for Naturalistic Evaluation of Large Audio-Language Models","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10736","citing_title":"BlasBench: An Open Benchmark for Irish Speech Recognition","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ","json":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ.json","graph_json":"https://pith.science/api/pith-number/RC2VCJ5PQGRMTU26BQJUDDUTCQ/graph.json","events_json":"https://pith.science/api/pith-number/RC2VCJ5PQGRMTU26BQJUDDUTCQ/events.json","paper":"https://pith.science/paper/RC2VCJ5P"},"agent_actions":{"view_html":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ","download_json":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ.json","view_paper":"https://pith.science/paper/RC2VCJ5P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.10752&json=true","fetch_graph":"https://pith.science/api/pith-number/RC2VCJ5PQGRMTU26BQJUDDUTCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/RC2VCJ5PQGRMTU26BQJUDDUTCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ/action/storage_attestation","attest_author":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ/action/author_attestation","sign_citation":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ/action/citation_signature","submit_replication":"https://pith.science/pith/RC2VCJ5PQGRMTU26BQJUDDUTCQ/action/replication_record"}},"created_at":"2026-07-05T04:14:28.134473+00:00","updated_at":"2026-07-05T04:14:28.134473+00:00"}