{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:TCF3APMDD26J7PNQZOZYZX3IGR","short_pith_number":"pith:TCF3APMD","schema_version":"1.0","canonical_sha256":"988bb03d831ebc9fbdb0cbb38cdf683479d75a3734304ecef76fdad44a929d3a","source":{"kind":"arxiv","id":"1912.03063","version":1},"attestation_state":"computed","paper":{"title":"Weak Supervision helps Emergence of Word-Object Alignment and improves Vision-Language Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"Christian Wolf (LIRIS), Corentin Kervadec (LIRIS), Grigory Antipov, Moez Baccouche","submitted_at":"2019-12-06T11:04:08Z","abstract_excerpt":"The large adoption of the self-attention (i.e. transformer model) and BERT-like training principles has recently resulted in a number of high performing models on a large panoply of vision-and-language problems (such as Visual Question Answering (VQA), image retrieval, etc.). In this paper we claim that these State-Of-The-Art (SOTA) approaches perform reasonably well in structuring information inside a single modality but, despite their impressive performances , they tend to struggle to identify fine-grained inter-modality relationships. Indeed, such relations are frequently assumed to be impl"},"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":"1912.03063","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-12-06T11:04:08Z","cross_cats_sorted":["cs.CL","cs.LG","cs.NE"],"title_canon_sha256":"519b23fdccbce00499d428298831a5e010ed38a923b1db136dbbd51ea5958107","abstract_canon_sha256":"29fad2caccc8b81f78571d10c30c249366ce44c579169516f402eaa5fd7089b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:24:30.150028Z","signature_b64":"8Scr9U0+7o7+fjO/0FFCAKNXvbBs7HHkD7rrZBRhUzL0a3mAsMrJBv7BWyPIJnosS+2dXgVijNNLVr+F7WvjBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"988bb03d831ebc9fbdb0cbb38cdf683479d75a3734304ecef76fdad44a929d3a","last_reissued_at":"2026-07-05T00:24:30.149608Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:24:30.149608Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Weak Supervision helps Emergence of Word-Object Alignment and improves Vision-Language Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"Christian Wolf (LIRIS), Corentin Kervadec (LIRIS), Grigory Antipov, Moez Baccouche","submitted_at":"2019-12-06T11:04:08Z","abstract_excerpt":"The large adoption of the self-attention (i.e. transformer model) and BERT-like training principles has recently resulted in a number of high performing models on a large panoply of vision-and-language problems (such as Visual Question Answering (VQA), image retrieval, etc.). In this paper we claim that these State-Of-The-Art (SOTA) approaches perform reasonably well in structuring information inside a single modality but, despite their impressive performances , they tend to struggle to identify fine-grained inter-modality relationships. Indeed, such relations are frequently assumed to be impl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.03063","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/1912.03063/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":"1912.03063","created_at":"2026-07-05T00:24:30.149671+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.03063v1","created_at":"2026-07-05T00:24:30.149671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.03063","created_at":"2026-07-05T00:24:30.149671+00:00"},{"alias_kind":"pith_short_12","alias_value":"TCF3APMDD26J","created_at":"2026-07-05T00:24:30.149671+00:00"},{"alias_kind":"pith_short_16","alias_value":"TCF3APMDD26J7PNQ","created_at":"2026-07-05T00:24:30.149671+00:00"},{"alias_kind":"pith_short_8","alias_value":"TCF3APMD","created_at":"2026-07-05T00:24:30.149671+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.02402","citing_title":"RG-SAN: Rule-Guided Spatial Awareness Network for End-to-End 3D Referring Expression Segmentation","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR","json":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR.json","graph_json":"https://pith.science/api/pith-number/TCF3APMDD26J7PNQZOZYZX3IGR/graph.json","events_json":"https://pith.science/api/pith-number/TCF3APMDD26J7PNQZOZYZX3IGR/events.json","paper":"https://pith.science/paper/TCF3APMD"},"agent_actions":{"view_html":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR","download_json":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR.json","view_paper":"https://pith.science/paper/TCF3APMD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.03063&json=true","fetch_graph":"https://pith.science/api/pith-number/TCF3APMDD26J7PNQZOZYZX3IGR/graph.json","fetch_events":"https://pith.science/api/pith-number/TCF3APMDD26J7PNQZOZYZX3IGR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR/action/storage_attestation","attest_author":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR/action/author_attestation","sign_citation":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR/action/citation_signature","submit_replication":"https://pith.science/pith/TCF3APMDD26J7PNQZOZYZX3IGR/action/replication_record"}},"created_at":"2026-07-05T00:24:30.149671+00:00","updated_at":"2026-07-05T00:24:30.149671+00:00"}