{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6GQUJ7AYHEEIDVQAVR5AQ64CA2","short_pith_number":"pith:6GQUJ7AY","schema_version":"1.0","canonical_sha256":"f1a144fc18390881d600ac7a087b82069013db6adbc60c90b61012ab6676822f","source":{"kind":"arxiv","id":"2606.20676","version":1},"attestation_state":"computed","paper":{"title":"Jury Duty: Calibration and Orientation Failures in MLLM-as-a-Judge Under Cultural Ambiguity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Andreas Vlachos, Daniel Lee, Eunkyu Park, Harsh Sharma, Jeonghwan Kim, Kah Mun Chia, Pranav Narayanan Venkit, Shafiq Joty","submitted_at":"2026-06-12T15:53:40Z","abstract_excerpt":"MLLM-as-a-Judge is conventionally validated by agreement with human annotations, but this metric is undefined when the human pool is culturally heterogeneous. We introduce VOIR DIRE, a multimodal benchmark of 626 culturally paired image--prompt artifacts spanning U.S. and mainland Chinese contexts across food, fashion, and architecture, with annotator pools that are within-pool reliable (a = 0.86/0.74) but cross-pool divergent on evaluation (Q1 r = -0.12). Across six MLLMs, the bias decomposes into two failures: a positivity-floor calibration failure (compressed scale use) and an orientation f"},"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":"2606.20676","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-12T15:53:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"f7e87ba14eb28abe6edb582d6bf3bbd038046d5ae4377b1e4f8835e54a30c33a","abstract_canon_sha256":"56d2ab5eeef4f706d6711da61e7a5b44407866b73a5127224f19672a18f2cc6e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:53.390311Z","signature_b64":"2l7vwpvnVz5CllBbAJOq+M0n2Htje1Ef8HV05A1kLHvPATyb7nkbIazX+16mRYd4P64tSRmvvkxGAwUSOo1DAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1a144fc18390881d600ac7a087b82069013db6adbc60c90b61012ab6676822f","last_reissued_at":"2026-06-23T00:11:53.389931Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:53.389931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Jury Duty: Calibration and Orientation Failures in MLLM-as-a-Judge Under Cultural Ambiguity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Andreas Vlachos, Daniel Lee, Eunkyu Park, Harsh Sharma, Jeonghwan Kim, Kah Mun Chia, Pranav Narayanan Venkit, Shafiq Joty","submitted_at":"2026-06-12T15:53:40Z","abstract_excerpt":"MLLM-as-a-Judge is conventionally validated by agreement with human annotations, but this metric is undefined when the human pool is culturally heterogeneous. We introduce VOIR DIRE, a multimodal benchmark of 626 culturally paired image--prompt artifacts spanning U.S. and mainland Chinese contexts across food, fashion, and architecture, with annotator pools that are within-pool reliable (a = 0.86/0.74) but cross-pool divergent on evaluation (Q1 r = -0.12). Across six MLLMs, the bias decomposes into two failures: a positivity-floor calibration failure (compressed scale use) and an orientation f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20676","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/2606.20676/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":"2606.20676","created_at":"2026-06-23T00:11:53.389993+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20676v1","created_at":"2026-06-23T00:11:53.389993+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20676","created_at":"2026-06-23T00:11:53.389993+00:00"},{"alias_kind":"pith_short_12","alias_value":"6GQUJ7AYHEEI","created_at":"2026-06-23T00:11:53.389993+00:00"},{"alias_kind":"pith_short_16","alias_value":"6GQUJ7AYHEEIDVQA","created_at":"2026-06-23T00:11:53.389993+00:00"},{"alias_kind":"pith_short_8","alias_value":"6GQUJ7AY","created_at":"2026-06-23T00:11:53.389993+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/6GQUJ7AYHEEIDVQAVR5AQ64CA2","json":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2.json","graph_json":"https://pith.science/api/pith-number/6GQUJ7AYHEEIDVQAVR5AQ64CA2/graph.json","events_json":"https://pith.science/api/pith-number/6GQUJ7AYHEEIDVQAVR5AQ64CA2/events.json","paper":"https://pith.science/paper/6GQUJ7AY"},"agent_actions":{"view_html":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2","download_json":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2.json","view_paper":"https://pith.science/paper/6GQUJ7AY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20676&json=true","fetch_graph":"https://pith.science/api/pith-number/6GQUJ7AYHEEIDVQAVR5AQ64CA2/graph.json","fetch_events":"https://pith.science/api/pith-number/6GQUJ7AYHEEIDVQAVR5AQ64CA2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2/action/storage_attestation","attest_author":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2/action/author_attestation","sign_citation":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2/action/citation_signature","submit_replication":"https://pith.science/pith/6GQUJ7AYHEEIDVQAVR5AQ64CA2/action/replication_record"}},"created_at":"2026-06-23T00:11:53.389993+00:00","updated_at":"2026-06-23T00:11:53.389993+00:00"}