{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DTFDUJEBX7UBGNDP5H5OUN4M6D","short_pith_number":"pith:DTFDUJEB","schema_version":"1.0","canonical_sha256":"1cca3a2481bfe813346fe9faea378cf0e69a009a09b9c5d6af3ac8341d747884","source":{"kind":"arxiv","id":"2508.16762","version":1},"attestation_state":"computed","paper":{"title":"Toward Socially Aware Vision-Language Models: Evaluating Cultural Competence Through Multimodal Story Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Arka Mukherjee, Shreya Ghosh","submitted_at":"2025-08-22T19:39:02Z","abstract_excerpt":"As Vision-Language Models (VLMs) achieve widespread deployment across diverse cultural contexts, ensuring their cultural competence becomes critical for responsible AI systems. While prior work has evaluated cultural awareness in text-only models and VLM object recognition tasks, no research has systematically assessed how VLMs adapt outputs when cultural identity cues are embedded in both textual prompts and visual inputs during generative tasks. We present the first comprehensive evaluation of VLM cultural competence through multimodal story generation, developing a novel multimodal framewor"},"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":"2508.16762","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-22T19:39:02Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"fdacc2133985f57bc577f27c2a0f617f2e6d11fae6fbf4eb7943d68c95117b79","abstract_canon_sha256":"4f3c23257ad3b464dc74708cc5beb34d064578286946214e1d54e8e4aee80074"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:58:03.193124Z","signature_b64":"3dkUOaBCSEMfa7fJSqo29hJbp6IbiVAl5k26kpcqyTlFoxNUvat39KdPcdsmsZziiueKY4voT+JUML9fYPsCBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1cca3a2481bfe813346fe9faea378cf0e69a009a09b9c5d6af3ac8341d747884","last_reissued_at":"2026-07-05T11:58:03.192689Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:58:03.192689Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Toward Socially Aware Vision-Language Models: Evaluating Cultural Competence Through Multimodal Story Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Arka Mukherjee, Shreya Ghosh","submitted_at":"2025-08-22T19:39:02Z","abstract_excerpt":"As Vision-Language Models (VLMs) achieve widespread deployment across diverse cultural contexts, ensuring their cultural competence becomes critical for responsible AI systems. While prior work has evaluated cultural awareness in text-only models and VLM object recognition tasks, no research has systematically assessed how VLMs adapt outputs when cultural identity cues are embedded in both textual prompts and visual inputs during generative tasks. We present the first comprehensive evaluation of VLM cultural competence through multimodal story generation, developing a novel multimodal framewor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.16762","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/2508.16762/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":"2508.16762","created_at":"2026-07-05T11:58:03.192746+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.16762v1","created_at":"2026-07-05T11:58:03.192746+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.16762","created_at":"2026-07-05T11:58:03.192746+00:00"},{"alias_kind":"pith_short_12","alias_value":"DTFDUJEBX7UB","created_at":"2026-07-05T11:58:03.192746+00:00"},{"alias_kind":"pith_short_16","alias_value":"DTFDUJEBX7UBGNDP","created_at":"2026-07-05T11:58:03.192746+00:00"},{"alias_kind":"pith_short_8","alias_value":"DTFDUJEB","created_at":"2026-07-05T11:58:03.192746+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/DTFDUJEBX7UBGNDP5H5OUN4M6D","json":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D.json","graph_json":"https://pith.science/api/pith-number/DTFDUJEBX7UBGNDP5H5OUN4M6D/graph.json","events_json":"https://pith.science/api/pith-number/DTFDUJEBX7UBGNDP5H5OUN4M6D/events.json","paper":"https://pith.science/paper/DTFDUJEB"},"agent_actions":{"view_html":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D","download_json":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D.json","view_paper":"https://pith.science/paper/DTFDUJEB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.16762&json=true","fetch_graph":"https://pith.science/api/pith-number/DTFDUJEBX7UBGNDP5H5OUN4M6D/graph.json","fetch_events":"https://pith.science/api/pith-number/DTFDUJEBX7UBGNDP5H5OUN4M6D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D/action/storage_attestation","attest_author":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D/action/author_attestation","sign_citation":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D/action/citation_signature","submit_replication":"https://pith.science/pith/DTFDUJEBX7UBGNDP5H5OUN4M6D/action/replication_record"}},"created_at":"2026-07-05T11:58:03.192746+00:00","updated_at":"2026-07-05T11:58:03.192746+00:00"}