{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:VKGOPENVTUCCRCC4NGNE4VNNM5","short_pith_number":"pith:VKGOPENV","schema_version":"1.0","canonical_sha256":"aa8ce791b59d0428885c699a4e55ad67600b5197b926d2b8ba76c9326eca3e12","source":{"kind":"arxiv","id":"2306.15162","version":2},"attestation_state":"computed","paper":{"title":"YouTube-ASL: A Large-Scale, Open-Domain American Sign Language-English Parallel Corpus","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"David Uthus, Garrett Tanzer, Manfred Georg","submitted_at":"2023-06-27T02:44:07Z","abstract_excerpt":"Machine learning for sign languages is bottlenecked by data. In this paper, we present YouTube-ASL, a large-scale, open-domain corpus of American Sign Language (ASL) videos and accompanying English captions drawn from YouTube. With ~1000 hours of videos and >2500 unique signers, YouTube-ASL is ~3x as large and has ~10x as many unique signers as the largest prior ASL dataset. We train baseline models for ASL to English translation on YouTube-ASL and evaluate them on How2Sign, where we achieve a new finetuned state of the art of 12.39 BLEU and, for the first time, report zero-shot results."},"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":"2306.15162","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-27T02:44:07Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"297d722512350ddea77c79733a4d886011421abb8081674edb520a826a4598f4","abstract_canon_sha256":"666d9d5f66a2952cb49ec283800a43c4d94fab13933a5dbd442c863d3f35bbd8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:05:45.598465Z","signature_b64":"57SP7MlYFIf2O3SMsDV3Dh85sE26WnrIK7KsIrfjLJKgds0jffIPRFOcZ9suEjSdmPugNA2obsXKR4xdzuziBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa8ce791b59d0428885c699a4e55ad67600b5197b926d2b8ba76c9326eca3e12","last_reissued_at":"2026-07-05T07:05:45.597997Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:05:45.597997Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"YouTube-ASL: A Large-Scale, Open-Domain American Sign Language-English Parallel Corpus","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"David Uthus, Garrett Tanzer, Manfred Georg","submitted_at":"2023-06-27T02:44:07Z","abstract_excerpt":"Machine learning for sign languages is bottlenecked by data. In this paper, we present YouTube-ASL, a large-scale, open-domain corpus of American Sign Language (ASL) videos and accompanying English captions drawn from YouTube. With ~1000 hours of videos and >2500 unique signers, YouTube-ASL is ~3x as large and has ~10x as many unique signers as the largest prior ASL dataset. We train baseline models for ASL to English translation on YouTube-ASL and evaluate them on How2Sign, where we achieve a new finetuned state of the art of 12.39 BLEU and, for the first time, report zero-shot results."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.15162","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/2306.15162/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":"2306.15162","created_at":"2026-07-05T07:05:45.598064+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.15162v2","created_at":"2026-07-05T07:05:45.598064+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.15162","created_at":"2026-07-05T07:05:45.598064+00:00"},{"alias_kind":"pith_short_12","alias_value":"VKGOPENVTUCC","created_at":"2026-07-05T07:05:45.598064+00:00"},{"alias_kind":"pith_short_16","alias_value":"VKGOPENVTUCCRCC4","created_at":"2026-07-05T07:05:45.598064+00:00"},{"alias_kind":"pith_short_8","alias_value":"VKGOPENV","created_at":"2026-07-05T07:05:45.598064+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.15301","citing_title":"Think in Latent Thoughts: A New Paradigm for Gloss-Free Sign Language Translation","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5","json":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5.json","graph_json":"https://pith.science/api/pith-number/VKGOPENVTUCCRCC4NGNE4VNNM5/graph.json","events_json":"https://pith.science/api/pith-number/VKGOPENVTUCCRCC4NGNE4VNNM5/events.json","paper":"https://pith.science/paper/VKGOPENV"},"agent_actions":{"view_html":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5","download_json":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5.json","view_paper":"https://pith.science/paper/VKGOPENV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.15162&json=true","fetch_graph":"https://pith.science/api/pith-number/VKGOPENVTUCCRCC4NGNE4VNNM5/graph.json","fetch_events":"https://pith.science/api/pith-number/VKGOPENVTUCCRCC4NGNE4VNNM5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5/action/storage_attestation","attest_author":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5/action/author_attestation","sign_citation":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5/action/citation_signature","submit_replication":"https://pith.science/pith/VKGOPENVTUCCRCC4NGNE4VNNM5/action/replication_record"}},"created_at":"2026-07-05T07:05:45.598064+00:00","updated_at":"2026-07-05T07:05:45.598064+00:00"}