{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:7AOC2LOTXAHNLTOBB4JC4VOM43","short_pith_number":"pith:7AOC2LOT","schema_version":"1.0","canonical_sha256":"f81c2d2dd3b80ed5cdc10f122e55cce6cf7f6c5ff054d63aeb8a2c50b78e4532","source":{"kind":"arxiv","id":"2312.07395","version":2},"attestation_state":"computed","paper":{"title":"A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Aida Nematzadeh, Andrew Zisserman, Antoine Miech, Jiajun Shen, Joe Heyward, Justin Chiu, Pinelopi Papalampidi, Shreya Pathak, Skanda Koppula, Viorica Patraucean","submitted_at":"2023-12-12T16:10:19Z","abstract_excerpt":"Understanding long, real-world videos requires modeling of long-range visual dependencies. To this end, we explore video-first architectures, building on the common paradigm of transferring large-scale, image--text models to video via shallow temporal fusion. However, we expose two limitations to the approach: (1) decreased spatial capabilities, likely due to poor video--language alignment in standard video datasets, and (2) higher memory consumption, bottlenecking the number of frames that can be processed. To mitigate the memory bottleneck, we systematically analyze the memory/accuracy trade"},"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":"2312.07395","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-12-12T16:10:19Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"426faf8a2ca3fe6804cb675a1e9213360f161167aab8f6111900dfa10a25d355","abstract_canon_sha256":"4ad2bf4a6d515d5dd2fa7959776446def580639ea9857c2f52e02a262289b93b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:55:07.498811Z","signature_b64":"GH8rdWNuG0aYV8ccadXHi8YdTOL7xIupwrC/XBCaaWK22Da0wZzY6Tvm5hULCoUfSyUVR0KTx57gzWRikZ9JAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f81c2d2dd3b80ed5cdc10f122e55cce6cf7f6c5ff054d63aeb8a2c50b78e4532","last_reissued_at":"2026-07-05T09:55:07.498233Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:55:07.498233Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Aida Nematzadeh, Andrew Zisserman, Antoine Miech, Jiajun Shen, Joe Heyward, Justin Chiu, Pinelopi Papalampidi, Shreya Pathak, Skanda Koppula, Viorica Patraucean","submitted_at":"2023-12-12T16:10:19Z","abstract_excerpt":"Understanding long, real-world videos requires modeling of long-range visual dependencies. To this end, we explore video-first architectures, building on the common paradigm of transferring large-scale, image--text models to video via shallow temporal fusion. However, we expose two limitations to the approach: (1) decreased spatial capabilities, likely due to poor video--language alignment in standard video datasets, and (2) higher memory consumption, bottlenecking the number of frames that can be processed. To mitigate the memory bottleneck, we systematically analyze the memory/accuracy trade"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.07395","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/2312.07395/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":"2312.07395","created_at":"2026-07-05T09:55:07.498299+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.07395v2","created_at":"2026-07-05T09:55:07.498299+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.07395","created_at":"2026-07-05T09:55:07.498299+00:00"},{"alias_kind":"pith_short_12","alias_value":"7AOC2LOTXAHN","created_at":"2026-07-05T09:55:07.498299+00:00"},{"alias_kind":"pith_short_16","alias_value":"7AOC2LOTXAHNLTOB","created_at":"2026-07-05T09:55:07.498299+00:00"},{"alias_kind":"pith_short_8","alias_value":"7AOC2LOT","created_at":"2026-07-05T09:55:07.498299+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/7AOC2LOTXAHNLTOBB4JC4VOM43","json":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43.json","graph_json":"https://pith.science/api/pith-number/7AOC2LOTXAHNLTOBB4JC4VOM43/graph.json","events_json":"https://pith.science/api/pith-number/7AOC2LOTXAHNLTOBB4JC4VOM43/events.json","paper":"https://pith.science/paper/7AOC2LOT"},"agent_actions":{"view_html":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43","download_json":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43.json","view_paper":"https://pith.science/paper/7AOC2LOT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.07395&json=true","fetch_graph":"https://pith.science/api/pith-number/7AOC2LOTXAHNLTOBB4JC4VOM43/graph.json","fetch_events":"https://pith.science/api/pith-number/7AOC2LOTXAHNLTOBB4JC4VOM43/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43/action/storage_attestation","attest_author":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43/action/author_attestation","sign_citation":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43/action/citation_signature","submit_replication":"https://pith.science/pith/7AOC2LOTXAHNLTOBB4JC4VOM43/action/replication_record"}},"created_at":"2026-07-05T09:55:07.498299+00:00","updated_at":"2026-07-05T09:55:07.498299+00:00"}