{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TZMQDXTNDOUKFRXZINTHLHWVAK","short_pith_number":"pith:TZMQDXTN","schema_version":"1.0","canonical_sha256":"9e5901de6d1ba8a2c6f94366759ed50290e896d16b12eccdf8b3481fe30dbb0a","source":{"kind":"arxiv","id":"2102.02779","version":2},"attestation_state":"computed","paper":{"title":"Unifying Vision-and-Language Tasks via Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hao Tan, Jaemin Cho, Jie Lei, Mohit Bansal","submitted_at":"2021-02-04T17:59:30Z","abstract_excerpt":"Existing methods for vision-and-language learning typically require designing task-specific architectures and objectives for each task. For example, a multi-label answer classifier for visual question answering, a region scorer for referring expression comprehension, and a language decoder for image captioning, etc. To alleviate these hassles, in this work, we propose a unified framework that learns different tasks in a single architecture with the same language modeling objective, i.e., multimodal conditional text generation, where our models learn to generate labels in text based on the visu"},"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":"2102.02779","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-04T17:59:30Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"30bec728885e6481746fa29d3662a7c44f8571b3ddafc3d32a9185fdbe01165b","abstract_canon_sha256":"20656c1642c21748ec3f37a04203fb03e90a062da434ade908f70c667f37ff1f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:42:29.322508Z","signature_b64":"kpM83bEr6WRTxyo4fvGmt0H2/k6FewvqU0shc36y8EfusCrTJFuO+xjbId/umf12j2bYTj8bIzQq3Xo2T2b/Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e5901de6d1ba8a2c6f94366759ed50290e896d16b12eccdf8b3481fe30dbb0a","last_reissued_at":"2026-07-05T02:42:29.321977Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:42:29.321977Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unifying Vision-and-Language Tasks via Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hao Tan, Jaemin Cho, Jie Lei, Mohit Bansal","submitted_at":"2021-02-04T17:59:30Z","abstract_excerpt":"Existing methods for vision-and-language learning typically require designing task-specific architectures and objectives for each task. For example, a multi-label answer classifier for visual question answering, a region scorer for referring expression comprehension, and a language decoder for image captioning, etc. To alleviate these hassles, in this work, we propose a unified framework that learns different tasks in a single architecture with the same language modeling objective, i.e., multimodal conditional text generation, where our models learn to generate labels in text based on the visu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.02779","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/2102.02779/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":"2102.02779","created_at":"2026-07-05T02:42:29.322042+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.02779v2","created_at":"2026-07-05T02:42:29.322042+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.02779","created_at":"2026-07-05T02:42:29.322042+00:00"},{"alias_kind":"pith_short_12","alias_value":"TZMQDXTNDOUK","created_at":"2026-07-05T02:42:29.322042+00:00"},{"alias_kind":"pith_short_16","alias_value":"TZMQDXTNDOUKFRXZ","created_at":"2026-07-05T02:42:29.322042+00:00"},{"alias_kind":"pith_short_8","alias_value":"TZMQDXTN","created_at":"2026-07-05T02:42:29.322042+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2305.06500","citing_title":"InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2301.12597","citing_title":"BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14779","citing_title":"AIM: Asymmetric Information Masking for Visual Question Answering Continual Learning","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK","json":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK.json","graph_json":"https://pith.science/api/pith-number/TZMQDXTNDOUKFRXZINTHLHWVAK/graph.json","events_json":"https://pith.science/api/pith-number/TZMQDXTNDOUKFRXZINTHLHWVAK/events.json","paper":"https://pith.science/paper/TZMQDXTN"},"agent_actions":{"view_html":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK","download_json":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK.json","view_paper":"https://pith.science/paper/TZMQDXTN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.02779&json=true","fetch_graph":"https://pith.science/api/pith-number/TZMQDXTNDOUKFRXZINTHLHWVAK/graph.json","fetch_events":"https://pith.science/api/pith-number/TZMQDXTNDOUKFRXZINTHLHWVAK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK/action/storage_attestation","attest_author":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK/action/author_attestation","sign_citation":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK/action/citation_signature","submit_replication":"https://pith.science/pith/TZMQDXTNDOUKFRXZINTHLHWVAK/action/replication_record"}},"created_at":"2026-07-05T02:42:29.322042+00:00","updated_at":"2026-07-05T02:42:29.322042+00:00"}