{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TSB6YQWBZYNFAPJ3RSI2XFBIYU","short_pith_number":"pith:TSB6YQWB","schema_version":"1.0","canonical_sha256":"9c83ec42c1ce1a503d3b8c91ab9428c50d9c95818a3a07f18193285cb9499e18","source":{"kind":"arxiv","id":"2503.16826","version":1},"attestation_state":"computed","paper":{"title":"When Tom Eats Kimchi: Evaluating Cultural Bias of Multimodal Large Language Models in Cultural Mixture Contexts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alice Oh, Eunsu Kim, Huzama Ahmad, James Thorne, Javad Ismayilzada, Jun Seong Kim, Junyeong Park, Kyaw Ye Thu, Na Min An","submitted_at":"2025-03-21T03:50:05Z","abstract_excerpt":"In a highly globalized world, it is important for multi-modal large language models (MLLMs) to recognize and respond correctly to mixed-cultural inputs. For example, a model should correctly identify kimchi (Korean food) in an image both when an Asian woman is eating it, as well as an African man is eating it. However, current MLLMs show an over-reliance on the visual features of the person, leading to misclassification of the entities. To examine the robustness of MLLMs to different ethnicity, we introduce MixCuBe, a cross-cultural bias benchmark, and study elements from five countries and fo"},"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":"2503.16826","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-21T03:50:05Z","cross_cats_sorted":[],"title_canon_sha256":"5a0b4420fd94fc02d3fa6b418ee54b337d3caf08b077a9606a16908343127816","abstract_canon_sha256":"6162e9ecd24226f5fcc036ef729ce9e9ef07ea60a52f9be382cbee4f21ab7010"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:36:50.852217Z","signature_b64":"0Do04foQUMfmM5Vy8151ozYAx2VLz1mbI6psdJZLbTVb3H9MQPkPAmRHsnWCxJ4a5hVKsCWjMCamkfswgbYvDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c83ec42c1ce1a503d3b8c91ab9428c50d9c95818a3a07f18193285cb9499e18","last_reissued_at":"2026-07-05T10:36:50.851786Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:36:50.851786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Tom Eats Kimchi: Evaluating Cultural Bias of Multimodal Large Language Models in Cultural Mixture Contexts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alice Oh, Eunsu Kim, Huzama Ahmad, James Thorne, Javad Ismayilzada, Jun Seong Kim, Junyeong Park, Kyaw Ye Thu, Na Min An","submitted_at":"2025-03-21T03:50:05Z","abstract_excerpt":"In a highly globalized world, it is important for multi-modal large language models (MLLMs) to recognize and respond correctly to mixed-cultural inputs. For example, a model should correctly identify kimchi (Korean food) in an image both when an Asian woman is eating it, as well as an African man is eating it. However, current MLLMs show an over-reliance on the visual features of the person, leading to misclassification of the entities. To examine the robustness of MLLMs to different ethnicity, we introduce MixCuBe, a cross-cultural bias benchmark, and study elements from five countries and fo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.16826","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/2503.16826/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":"2503.16826","created_at":"2026-07-05T10:36:50.851852+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.16826v1","created_at":"2026-07-05T10:36:50.851852+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.16826","created_at":"2026-07-05T10:36:50.851852+00:00"},{"alias_kind":"pith_short_12","alias_value":"TSB6YQWBZYNF","created_at":"2026-07-05T10:36:50.851852+00:00"},{"alias_kind":"pith_short_16","alias_value":"TSB6YQWBZYNFAPJ3","created_at":"2026-07-05T10:36:50.851852+00:00"},{"alias_kind":"pith_short_8","alias_value":"TSB6YQWB","created_at":"2026-07-05T10:36:50.851852+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.20901","citing_title":"A Stereotype Content Analysis on Color-related Social Bias in Large Vision Language Models","ref_index":31,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU","json":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU.json","graph_json":"https://pith.science/api/pith-number/TSB6YQWBZYNFAPJ3RSI2XFBIYU/graph.json","events_json":"https://pith.science/api/pith-number/TSB6YQWBZYNFAPJ3RSI2XFBIYU/events.json","paper":"https://pith.science/paper/TSB6YQWB"},"agent_actions":{"view_html":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU","download_json":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU.json","view_paper":"https://pith.science/paper/TSB6YQWB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.16826&json=true","fetch_graph":"https://pith.science/api/pith-number/TSB6YQWBZYNFAPJ3RSI2XFBIYU/graph.json","fetch_events":"https://pith.science/api/pith-number/TSB6YQWBZYNFAPJ3RSI2XFBIYU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU/action/storage_attestation","attest_author":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU/action/author_attestation","sign_citation":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU/action/citation_signature","submit_replication":"https://pith.science/pith/TSB6YQWBZYNFAPJ3RSI2XFBIYU/action/replication_record"}},"created_at":"2026-07-05T10:36:50.851852+00:00","updated_at":"2026-07-05T10:36:50.851852+00:00"}