{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CROECZPVM54SX4YDHIHI6HIVYX","short_pith_number":"pith:CROECZPV","schema_version":"1.0","canonical_sha256":"145c4165f567792bf3033a0e8f1d15c5d59a48044720cfbfee52f86989b33ef8","source":{"kind":"arxiv","id":"2508.07414","version":2},"attestation_state":"computed","paper":{"title":"Grounding Multilingual Multimodal LLMs With Cultural Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Graham Neubig, Jean de Dieu Nyandwi, Simran Khanuja, Yueqi Song","submitted_at":"2025-08-10T16:24:11Z","abstract_excerpt":"Multimodal Large Language Models excel in high-resource settings, but often misinterpret long-tail cultural entities and underperform in low-resource languages. To address this gap, we propose a data-centric approach that directly grounds MLLMs in cultural knowledge. Leveraging a large scale knowledge graph from Wikidata, we collect images that represent culturally significant entities, and generate synthetic multilingual visual question answering data. The resulting dataset, CulturalGround, comprises 22 million high-quality, culturally-rich VQA pairs spanning 42 countries and 39 languages. We"},"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":"2508.07414","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-10T16:24:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"45af3dd86c66a16babfea8e8b351c96913a564c5506d4776d8ed43e461200996","abstract_canon_sha256":"89d93ca6a5c53d52a9b5cf5443050f55ee24418a901239dce3b808a42f33bae1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:52:24.321308Z","signature_b64":"xVGhrPjThTUYIjsy0QNHUGP3xryr6Rg2LpZsHXg3+PPoUO8NZrL2y1ru/vBVnXu5K2+jD8MZnZ+1zjyYlCP8Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"145c4165f567792bf3033a0e8f1d15c5d59a48044720cfbfee52f86989b33ef8","last_reissued_at":"2026-07-05T11:52:24.320883Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:52:24.320883Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grounding Multilingual Multimodal LLMs With Cultural Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Graham Neubig, Jean de Dieu Nyandwi, Simran Khanuja, Yueqi Song","submitted_at":"2025-08-10T16:24:11Z","abstract_excerpt":"Multimodal Large Language Models excel in high-resource settings, but often misinterpret long-tail cultural entities and underperform in low-resource languages. To address this gap, we propose a data-centric approach that directly grounds MLLMs in cultural knowledge. Leveraging a large scale knowledge graph from Wikidata, we collect images that represent culturally significant entities, and generate synthetic multilingual visual question answering data. The resulting dataset, CulturalGround, comprises 22 million high-quality, culturally-rich VQA pairs spanning 42 countries and 39 languages. We"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07414","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/2508.07414/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":"2508.07414","created_at":"2026-07-05T11:52:24.320938+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07414v2","created_at":"2026-07-05T11:52:24.320938+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07414","created_at":"2026-07-05T11:52:24.320938+00:00"},{"alias_kind":"pith_short_12","alias_value":"CROECZPVM54S","created_at":"2026-07-05T11:52:24.320938+00:00"},{"alias_kind":"pith_short_16","alias_value":"CROECZPVM54SX4YD","created_at":"2026-07-05T11:52:24.320938+00:00"},{"alias_kind":"pith_short_8","alias_value":"CROECZPV","created_at":"2026-07-05T11:52:24.320938+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/CROECZPVM54SX4YDHIHI6HIVYX","json":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX.json","graph_json":"https://pith.science/api/pith-number/CROECZPVM54SX4YDHIHI6HIVYX/graph.json","events_json":"https://pith.science/api/pith-number/CROECZPVM54SX4YDHIHI6HIVYX/events.json","paper":"https://pith.science/paper/CROECZPV"},"agent_actions":{"view_html":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX","download_json":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX.json","view_paper":"https://pith.science/paper/CROECZPV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07414&json=true","fetch_graph":"https://pith.science/api/pith-number/CROECZPVM54SX4YDHIHI6HIVYX/graph.json","fetch_events":"https://pith.science/api/pith-number/CROECZPVM54SX4YDHIHI6HIVYX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX/action/storage_attestation","attest_author":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX/action/author_attestation","sign_citation":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX/action/citation_signature","submit_replication":"https://pith.science/pith/CROECZPVM54SX4YDHIHI6HIVYX/action/replication_record"}},"created_at":"2026-07-05T11:52:24.320938+00:00","updated_at":"2026-07-05T11:52:24.320938+00:00"}