{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FKOYVUCJYSOQOULXYZPSRECYBM","short_pith_number":"pith:FKOYVUCJ","schema_version":"1.0","canonical_sha256":"2a9d8ad049c49d075177c65f2890580b24832cda132fb048a0ed634ac3cb050f","source":{"kind":"arxiv","id":"2304.00719","version":1},"attestation_state":"computed","paper":{"title":"Multi-Modal Representation Learning with Text-Driven Soft Masks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bohyung Han, Jaeyoo Park","submitted_at":"2023-04-03T05:07:49Z","abstract_excerpt":"We propose a visual-linguistic representation learning approach within a self-supervised learning framework by introducing a new operation, loss, and data augmentation strategy. First, we generate diverse features for the image-text matching (ITM) task via soft-masking the regions in an image, which are most relevant to a certain word in the corresponding caption, instead of completely removing them. Since our framework relies only on image-caption pairs with no fine-grained annotations, we identify the relevant regions to each word by computing the word-conditional visual attention using mult"},"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":"2304.00719","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-04-03T05:07:49Z","cross_cats_sorted":[],"title_canon_sha256":"5842bae0a4212874937a524f108e19ce15d5fd783f21f8dec03fb63f4cfb8a0d","abstract_canon_sha256":"b9ae54f1c09482395ea6334fbd209991057ab6b33e1082564e5978a10c91ecdc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:57:21.376565Z","signature_b64":"ko6FPAQLKQX8BDcAKdttcPZ06TW3CaJkW7bpsFlG9r5ivk9JyBlfMTDji6rN2+MQxdfkhqHMWCvtWdFqCaWXCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a9d8ad049c49d075177c65f2890580b24832cda132fb048a0ed634ac3cb050f","last_reissued_at":"2026-07-05T05:57:21.376042Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:57:21.376042Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Modal Representation Learning with Text-Driven Soft Masks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bohyung Han, Jaeyoo Park","submitted_at":"2023-04-03T05:07:49Z","abstract_excerpt":"We propose a visual-linguistic representation learning approach within a self-supervised learning framework by introducing a new operation, loss, and data augmentation strategy. First, we generate diverse features for the image-text matching (ITM) task via soft-masking the regions in an image, which are most relevant to a certain word in the corresponding caption, instead of completely removing them. Since our framework relies only on image-caption pairs with no fine-grained annotations, we identify the relevant regions to each word by computing the word-conditional visual attention using mult"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.00719","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/2304.00719/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":"2304.00719","created_at":"2026-07-05T05:57:21.376105+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.00719v1","created_at":"2026-07-05T05:57:21.376105+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.00719","created_at":"2026-07-05T05:57:21.376105+00:00"},{"alias_kind":"pith_short_12","alias_value":"FKOYVUCJYSOQ","created_at":"2026-07-05T05:57:21.376105+00:00"},{"alias_kind":"pith_short_16","alias_value":"FKOYVUCJYSOQOULX","created_at":"2026-07-05T05:57:21.376105+00:00"},{"alias_kind":"pith_short_8","alias_value":"FKOYVUCJ","created_at":"2026-07-05T05:57:21.376105+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/FKOYVUCJYSOQOULXYZPSRECYBM","json":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM.json","graph_json":"https://pith.science/api/pith-number/FKOYVUCJYSOQOULXYZPSRECYBM/graph.json","events_json":"https://pith.science/api/pith-number/FKOYVUCJYSOQOULXYZPSRECYBM/events.json","paper":"https://pith.science/paper/FKOYVUCJ"},"agent_actions":{"view_html":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM","download_json":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM.json","view_paper":"https://pith.science/paper/FKOYVUCJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.00719&json=true","fetch_graph":"https://pith.science/api/pith-number/FKOYVUCJYSOQOULXYZPSRECYBM/graph.json","fetch_events":"https://pith.science/api/pith-number/FKOYVUCJYSOQOULXYZPSRECYBM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM/action/storage_attestation","attest_author":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM/action/author_attestation","sign_citation":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM/action/citation_signature","submit_replication":"https://pith.science/pith/FKOYVUCJYSOQOULXYZPSRECYBM/action/replication_record"}},"created_at":"2026-07-05T05:57:21.376105+00:00","updated_at":"2026-07-05T05:57:21.376105+00:00"}