{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VYHSGZL67OLNOYT4LIMAPJRDLB","short_pith_number":"pith:VYHSGZL6","schema_version":"1.0","canonical_sha256":"ae0f23657efb96d7627c5a1807a6235868df036b0fea996a2b617b81d124f0f1","source":{"kind":"arxiv","id":"2205.05019","version":2},"attestation_state":"computed","paper":{"title":"Learning to Answer Visual Questions from Web Videos","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Antoine Miech, Antoine Yang, Cordelia Schmid, Ivan Laptev, Josef Sivic","submitted_at":"2022-05-10T16:34:26Z","abstract_excerpt":"Recent methods for visual question answering rely on large-scale annotated datasets. Manual annotation of questions and answers for videos, however, is tedious, expensive and prevents scalability. In this work, we propose to avoid manual annotation and generate a large-scale training dataset for video question answering making use of automatic cross-modal supervision. We leverage a question generation transformer trained on text data and use it to generate question-answer pairs from transcribed video narrations. Given narrated videos, we then automatically generate the HowToVQA69M dataset with"},"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":"2205.05019","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-10T16:34:26Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"015f93d0cd5906e092e4282e320a606ef7f5d2e38eadaeeefcc26df75fcf25f9","abstract_canon_sha256":"63638b1286ce3b2486717becbd25df343a443cdc59d1bdd56e874bba2bfd10f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:22:22.235455Z","signature_b64":"jfC9NM1yId3HosB0LXNToo24ntMk3YbbM+eiavHjo8KMkU9xdA0xB4YsFX5n9/d5CWmPpoq7axQtwFSOARlTDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae0f23657efb96d7627c5a1807a6235868df036b0fea996a2b617b81d124f0f1","last_reissued_at":"2026-07-05T04:22:22.234951Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:22:22.234951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Answer Visual Questions from Web Videos","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Antoine Miech, Antoine Yang, Cordelia Schmid, Ivan Laptev, Josef Sivic","submitted_at":"2022-05-10T16:34:26Z","abstract_excerpt":"Recent methods for visual question answering rely on large-scale annotated datasets. Manual annotation of questions and answers for videos, however, is tedious, expensive and prevents scalability. In this work, we propose to avoid manual annotation and generate a large-scale training dataset for video question answering making use of automatic cross-modal supervision. We leverage a question generation transformer trained on text data and use it to generate question-answer pairs from transcribed video narrations. Given narrated videos, we then automatically generate the HowToVQA69M dataset with"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.05019","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/2205.05019/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":"2205.05019","created_at":"2026-07-05T04:22:22.235023+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.05019v2","created_at":"2026-07-05T04:22:22.235023+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.05019","created_at":"2026-07-05T04:22:22.235023+00:00"},{"alias_kind":"pith_short_12","alias_value":"VYHSGZL67OLN","created_at":"2026-07-05T04:22:22.235023+00:00"},{"alias_kind":"pith_short_16","alias_value":"VYHSGZL67OLNOYT4","created_at":"2026-07-05T04:22:22.235023+00:00"},{"alias_kind":"pith_short_8","alias_value":"VYHSGZL6","created_at":"2026-07-05T04:22:22.235023+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.20964","citing_title":"Hierarchical Banzhaf Interaction for General Video-Language Representation Learning","ref_index":81,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB","json":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB.json","graph_json":"https://pith.science/api/pith-number/VYHSGZL67OLNOYT4LIMAPJRDLB/graph.json","events_json":"https://pith.science/api/pith-number/VYHSGZL67OLNOYT4LIMAPJRDLB/events.json","paper":"https://pith.science/paper/VYHSGZL6"},"agent_actions":{"view_html":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB","download_json":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB.json","view_paper":"https://pith.science/paper/VYHSGZL6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.05019&json=true","fetch_graph":"https://pith.science/api/pith-number/VYHSGZL67OLNOYT4LIMAPJRDLB/graph.json","fetch_events":"https://pith.science/api/pith-number/VYHSGZL67OLNOYT4LIMAPJRDLB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB/action/storage_attestation","attest_author":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB/action/author_attestation","sign_citation":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB/action/citation_signature","submit_replication":"https://pith.science/pith/VYHSGZL67OLNOYT4LIMAPJRDLB/action/replication_record"}},"created_at":"2026-07-05T04:22:22.235023+00:00","updated_at":"2026-07-05T04:22:22.235023+00:00"}