{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:DYZIS3MHKY57ZYBCBHETM6ZSRJ","short_pith_number":"pith:DYZIS3MH","schema_version":"1.0","canonical_sha256":"1e32896d87563bfce02209c9367b328a519528e3233d1d2162a5d462e12540f1","source":{"kind":"arxiv","id":"2111.09296","version":3},"attestation_state":"computed","paper":{"title":"XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Alexei Baevski, Alexis Conneau, Andros Tjandra, Arun Babu, Changhan Wang, Juan Pino, Kritika Singh, Kushal Lakhotia, Michael Auli, Naman Goyal, Patrick von Platen, Qiantong Xu, Yatharth Saraf","submitted_at":"2021-11-17T18:49:42Z","abstract_excerpt":"This paper presents XLS-R, a large-scale model for cross-lingual speech representation learning based on wav2vec 2.0. We train models with up to 2B parameters on nearly half a million hours of publicly available speech audio in 128 languages, an order of magnitude more public data than the largest known prior work. Our evaluation covers a wide range of tasks, domains, data regimes and languages, both high and low-resource. On the CoVoST-2 speech translation benchmark, we improve the previous state of the art by an average of 7.4 BLEU over 21 translation directions into English. For speech reco"},"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":"2111.09296","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-11-17T18:49:42Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"bc6eb1af2f3ac5cf95c6964b5b50e9758ca48a29d124b3adf978b309b44360a9","abstract_canon_sha256":"a9a95f29cd8a9acf23dd66fef7faaffa7405acd3744e24b3982c4307f28e6c99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:41:27.872828Z","signature_b64":"NBDzUrmbToa6PfqpO0WnHXaSk07YEnImNC3EfR7npRXKJR1oW8Z5AwGAuzAn+q3YKOSffKEyfPTb2V+RWXXXAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e32896d87563bfce02209c9367b328a519528e3233d1d2162a5d462e12540f1","last_reissued_at":"2026-07-05T03:41:27.872433Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:41:27.872433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Alexei Baevski, Alexis Conneau, Andros Tjandra, Arun Babu, Changhan Wang, Juan Pino, Kritika Singh, Kushal Lakhotia, Michael Auli, Naman Goyal, Patrick von Platen, Qiantong Xu, Yatharth Saraf","submitted_at":"2021-11-17T18:49:42Z","abstract_excerpt":"This paper presents XLS-R, a large-scale model for cross-lingual speech representation learning based on wav2vec 2.0. We train models with up to 2B parameters on nearly half a million hours of publicly available speech audio in 128 languages, an order of magnitude more public data than the largest known prior work. Our evaluation covers a wide range of tasks, domains, data regimes and languages, both high and low-resource. On the CoVoST-2 speech translation benchmark, we improve the previous state of the art by an average of 7.4 BLEU over 21 translation directions into English. For speech reco"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.09296","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/2111.09296/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":"2111.09296","created_at":"2026-07-05T03:41:27.872492+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.09296v3","created_at":"2026-07-05T03:41:27.872492+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.09296","created_at":"2026-07-05T03:41:27.872492+00:00"},{"alias_kind":"pith_short_12","alias_value":"DYZIS3MHKY57","created_at":"2026-07-05T03:41:27.872492+00:00"},{"alias_kind":"pith_short_16","alias_value":"DYZIS3MHKY57ZYBC","created_at":"2026-07-05T03:41:27.872492+00:00"},{"alias_kind":"pith_short_8","alias_value":"DYZIS3MH","created_at":"2026-07-05T03:41:27.872492+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":21,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26384","citing_title":"What Do Deepfake Benchmarks Measure? An Audit Using Frozen Self-Supervised Representations","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22274","citing_title":"From Speech to Text Corpora: Evaluating ASR-Based Data Acquisition for Low-Resource Fongbe and Hausa","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21210","citing_title":"Impact Analysis of Speech Representation Learning Models for Acoustic Side-Channel Attack","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20478","citing_title":"Beyond Speaker Independence: Evaluating Cross-Lingual Acoustic-to-Articulatory Inversion Across Finnish and Russian","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19910","citing_title":"Light-weight Pronunciation Assessment via Discrete Speech Token Surprisal","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18659","citing_title":"Responsible ASR: Overcoming Challenges of Foundational Models in Narrow-Band and Low-Resource Settings","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11674","citing_title":"SpAArSIST: Sparsified AASIST for Efficient and Reliable Anti-Spoofing","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11542","citing_title":"Pretrained self-supervised speech models can recognize unseen consonants","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10223","citing_title":"Dual-Branch Gated Fusion for Open-Set Audio Deepfake Source Tracing","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00387","citing_title":"From Objectives to Applications: Aligning Architectural Biases in Audio Self-Supervised Learning","ref_index":93,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30780","citing_title":"Detecting Audio Deepfakes on the Edge:Lightweight SSL-Based Detection in a Browser Plugin","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30366","citing_title":"Escaping the Linearity Trap: Manifold Detours for Black-Box Adversarial Attacks on Singing Audio Deepfake Detection","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21210","citing_title":"Impact Analysis of Speech Representation Learning Models for Acoustic Side-Channel Attack","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27206","citing_title":"Syntactic Belief Update as the Driver of Garden Path Processing Difficulty","ref_index":295,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23201","citing_title":"MixFake: Benchmarking and Enhancing Audio Deepfake Detection in Diverse Real-world Mixed Audio","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17737","citing_title":"Profiling the Voice: Speaker-Specific Phoneme Fingerprinting for Speech Deepfake Detection","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2510.02864","citing_title":"Forensic Similarity for Speech Deepfakes","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2603.01482","citing_title":"A SUPERB-Style Benchmark of Self-Supervised Speech Models for Audio Deepfake Detection","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26057","citing_title":"Similarity Choice and Negative Scaling in Supervised Contrastive Learning for Deepfake Audio Detection","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13288","citing_title":"Giving Voice to the Constitution: Low-Resource Text-to-Speech for Quechua and Spanish Using a Bilingual Legal Corpus","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04598","citing_title":"Benchmarking Multilingual Speech Models on Pashto: Zero-Shot ASR, Script Failure, and Cross-Domain Evaluation","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ","json":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ.json","graph_json":"https://pith.science/api/pith-number/DYZIS3MHKY57ZYBCBHETM6ZSRJ/graph.json","events_json":"https://pith.science/api/pith-number/DYZIS3MHKY57ZYBCBHETM6ZSRJ/events.json","paper":"https://pith.science/paper/DYZIS3MH"},"agent_actions":{"view_html":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ","download_json":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ.json","view_paper":"https://pith.science/paper/DYZIS3MH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.09296&json=true","fetch_graph":"https://pith.science/api/pith-number/DYZIS3MHKY57ZYBCBHETM6ZSRJ/graph.json","fetch_events":"https://pith.science/api/pith-number/DYZIS3MHKY57ZYBCBHETM6ZSRJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ/action/storage_attestation","attest_author":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ/action/author_attestation","sign_citation":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ/action/citation_signature","submit_replication":"https://pith.science/pith/DYZIS3MHKY57ZYBCBHETM6ZSRJ/action/replication_record"}},"created_at":"2026-07-05T03:41:27.872492+00:00","updated_at":"2026-07-05T03:41:27.872492+00:00"}