{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:I6AHPNSRHOXIICW7Y4K5DAN2BN","short_pith_number":"pith:I6AHPNSR","schema_version":"1.0","canonical_sha256":"478077b6513bae840adfc715d181ba0b7d4927694b8ab7d1ebc723901788d634","source":{"kind":"arxiv","id":"2501.01282","version":1},"attestation_state":"computed","paper":{"title":"CultureVLM: Characterizing and Improving Cultural Understanding of Vision-Language Models for over 100 Countries","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Cheng Li, Derek F. Wong, Haipeng Chen, Jindong Wang, Lichao Sun, Qingsong Wen, Shudong Liu, Xing Xie, Yiqiao Jin","submitted_at":"2025-01-02T14:42:37Z","abstract_excerpt":"Vision-language models (VLMs) have advanced human-AI interaction but struggle with cultural understanding, often misinterpreting symbols, gestures, and artifacts due to biases in predominantly Western-centric training data. In this paper, we construct CultureVerse, a large-scale multimodal benchmark covering 19, 682 cultural concepts, 188 countries/regions, 15 cultural concepts, and 3 question types, with the aim of characterizing and improving VLMs' multicultural understanding capabilities. Then, we propose CultureVLM, a series of VLMs fine-tuned on our dataset to achieve significant performa"},"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":"2501.01282","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-01-02T14:42:37Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"1f66cce1ac326a74d563ea7986e8edb9061674453f0cc915ab23494e56efa28d","abstract_canon_sha256":"44ab5805e1303117795239cc7ab7dc677a3e86834cd24969bacc82280480040e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:56:17.395385Z","signature_b64":"GkYAdDeXY/bwZpc0CMeJcGlwSIQ9pr2hW7EDR9R0SjGVG19uzFZYkg5LnNJDFKOtwkMBoJN9by/N/INOyQz+Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"478077b6513bae840adfc715d181ba0b7d4927694b8ab7d1ebc723901788d634","last_reissued_at":"2026-07-05T09:56:17.394910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:56:17.394910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CultureVLM: Characterizing and Improving Cultural Understanding of Vision-Language Models for over 100 Countries","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Cheng Li, Derek F. Wong, Haipeng Chen, Jindong Wang, Lichao Sun, Qingsong Wen, Shudong Liu, Xing Xie, Yiqiao Jin","submitted_at":"2025-01-02T14:42:37Z","abstract_excerpt":"Vision-language models (VLMs) have advanced human-AI interaction but struggle with cultural understanding, often misinterpreting symbols, gestures, and artifacts due to biases in predominantly Western-centric training data. In this paper, we construct CultureVerse, a large-scale multimodal benchmark covering 19, 682 cultural concepts, 188 countries/regions, 15 cultural concepts, and 3 question types, with the aim of characterizing and improving VLMs' multicultural understanding capabilities. Then, we propose CultureVLM, a series of VLMs fine-tuned on our dataset to achieve significant performa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.01282","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/2501.01282/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":"2501.01282","created_at":"2026-07-05T09:56:17.394966+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.01282v1","created_at":"2026-07-05T09:56:17.394966+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.01282","created_at":"2026-07-05T09:56:17.394966+00:00"},{"alias_kind":"pith_short_12","alias_value":"I6AHPNSRHOXI","created_at":"2026-07-05T09:56:17.394966+00:00"},{"alias_kind":"pith_short_16","alias_value":"I6AHPNSRHOXIICW7","created_at":"2026-07-05T09:56:17.394966+00:00"},{"alias_kind":"pith_short_8","alias_value":"I6AHPNSR","created_at":"2026-07-05T09:56:17.394966+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30800","citing_title":"Computer-Aided Tagging on Wikimedia Commons: Designing for Human-AI Collaboration in Open Knowledge Work","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16716","citing_title":"When Cultures Move: Measuring and Improving Multicultural Text-to-Video Generation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2503.21460","citing_title":"Large Language Model Agent: A Survey on Methodology, Applications and Challenges","ref_index":248,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16716","citing_title":"When Cultures Move: Measuring and Improving Multicultural Text-to-Video Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18423","citing_title":"BhashaSutra: A Task-Centric Unified Survey of Indian NLP Datasets, Corpora, and Resources","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07338","citing_title":"Appear2Meaning: A Cross-Cultural Benchmark for Structured Cultural Metadata Inference from Images","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN","json":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN.json","graph_json":"https://pith.science/api/pith-number/I6AHPNSRHOXIICW7Y4K5DAN2BN/graph.json","events_json":"https://pith.science/api/pith-number/I6AHPNSRHOXIICW7Y4K5DAN2BN/events.json","paper":"https://pith.science/paper/I6AHPNSR"},"agent_actions":{"view_html":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN","download_json":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN.json","view_paper":"https://pith.science/paper/I6AHPNSR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.01282&json=true","fetch_graph":"https://pith.science/api/pith-number/I6AHPNSRHOXIICW7Y4K5DAN2BN/graph.json","fetch_events":"https://pith.science/api/pith-number/I6AHPNSRHOXIICW7Y4K5DAN2BN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN/action/storage_attestation","attest_author":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN/action/author_attestation","sign_citation":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN/action/citation_signature","submit_replication":"https://pith.science/pith/I6AHPNSRHOXIICW7Y4K5DAN2BN/action/replication_record"}},"created_at":"2026-07-05T09:56:17.394966+00:00","updated_at":"2026-07-05T09:56:17.394966+00:00"}