{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TMVQOUXJKXDM56Q5KHLBGM3S2A","short_pith_number":"pith:TMVQOUXJ","schema_version":"1.0","canonical_sha256":"9b2b0752e955c6cefa1d51d6133372d00a2ad658bf7413ab9cff5c69b6239058","source":{"kind":"arxiv","id":"2206.07160","version":1},"attestation_state":"computed","paper":{"title":"LAVENDER: Unifying Video-Language Understanding as Masked Language Modeling","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ce Liu, Chung-Ching Lin, Kevin Lin, Lijuan Wang, Linjie Li, Zhe Gan, Zicheng Liu","submitted_at":"2022-06-14T20:43:25Z","abstract_excerpt":"Unified vision-language frameworks have greatly advanced in recent years, most of which adopt an encoder-decoder architecture to unify image-text tasks as sequence-to-sequence generation. However, existing video-language (VidL) models still require task-specific designs in model architecture and training objectives for each task. In this work, we explore a unified VidL framework LAVENDER, where Masked Language Modeling (MLM) is used as the common interface for all pre-training and downstream tasks. Such unification leads to a simplified model architecture, where only a lightweight MLM head, in"},"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":"2206.07160","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2022-06-14T20:43:25Z","cross_cats_sorted":[],"title_canon_sha256":"aae7c82716a64a639cbf6be062aede09f8ad962d8b791aad981da4fb1c0bcc20","abstract_canon_sha256":"d512af4a784b16d6c469c493373e0d9e2f29f2f4fb3be69adbfa892917d1b38a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:31:50.510110Z","signature_b64":"42VHUf8LhEROTKW/gPlqAmnEZhUT62/bJQqdfXPIJk24yc1e86kquly3yAPKCz2RQT+o+153ishFb+13sw9pCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b2b0752e955c6cefa1d51d6133372d00a2ad658bf7413ab9cff5c69b6239058","last_reissued_at":"2026-07-05T04:31:50.509711Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:31:50.509711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LAVENDER: Unifying Video-Language Understanding as Masked Language Modeling","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ce Liu, Chung-Ching Lin, Kevin Lin, Lijuan Wang, Linjie Li, Zhe Gan, Zicheng Liu","submitted_at":"2022-06-14T20:43:25Z","abstract_excerpt":"Unified vision-language frameworks have greatly advanced in recent years, most of which adopt an encoder-decoder architecture to unify image-text tasks as sequence-to-sequence generation. However, existing video-language (VidL) models still require task-specific designs in model architecture and training objectives for each task. In this work, we explore a unified VidL framework LAVENDER, where Masked Language Modeling (MLM) is used as the common interface for all pre-training and downstream tasks. Such unification leads to a simplified model architecture, where only a lightweight MLM head, in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.07160","kind":"arxiv","version":1},"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/2206.07160/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":"2206.07160","created_at":"2026-07-05T04:31:50.509764+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.07160v1","created_at":"2026-07-05T04:31:50.509764+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.07160","created_at":"2026-07-05T04:31:50.509764+00:00"},{"alias_kind":"pith_short_12","alias_value":"TMVQOUXJKXDM","created_at":"2026-07-05T04:31:50.509764+00:00"},{"alias_kind":"pith_short_16","alias_value":"TMVQOUXJKXDM56Q5","created_at":"2026-07-05T04:31:50.509764+00:00"},{"alias_kind":"pith_short_8","alias_value":"TMVQOUXJ","created_at":"2026-07-05T04:31:50.509764+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2212.03191","citing_title":"InternVideo: General Video Foundation Models via Generative and Discriminative Learning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2307.06942","citing_title":"InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06355","citing_title":"VideoChat: Chat-Centric Video Understanding","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A","json":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A.json","graph_json":"https://pith.science/api/pith-number/TMVQOUXJKXDM56Q5KHLBGM3S2A/graph.json","events_json":"https://pith.science/api/pith-number/TMVQOUXJKXDM56Q5KHLBGM3S2A/events.json","paper":"https://pith.science/paper/TMVQOUXJ"},"agent_actions":{"view_html":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A","download_json":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A.json","view_paper":"https://pith.science/paper/TMVQOUXJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.07160&json=true","fetch_graph":"https://pith.science/api/pith-number/TMVQOUXJKXDM56Q5KHLBGM3S2A/graph.json","fetch_events":"https://pith.science/api/pith-number/TMVQOUXJKXDM56Q5KHLBGM3S2A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A/action/storage_attestation","attest_author":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A/action/author_attestation","sign_citation":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A/action/citation_signature","submit_replication":"https://pith.science/pith/TMVQOUXJKXDM56Q5KHLBGM3S2A/action/replication_record"}},"created_at":"2026-07-05T04:31:50.509764+00:00","updated_at":"2026-07-05T04:31:50.509764+00:00"}