{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FCKUV6WWHGSM2BG5PVTFY7MAQM","short_pith_number":"pith:FCKUV6WW","schema_version":"1.0","canonical_sha256":"28954afad639a4cd04dd7d665c7d808329cd4796a5d99f8df5a1dc266c27e86d","source":{"kind":"arxiv","id":"2505.12201","version":1},"attestation_state":"computed","paper":{"title":"How Reliable is Multilingual LLM-as-a-Judge?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Wei Liu, Xiyan Fu","submitted_at":"2025-05-18T02:32:35Z","abstract_excerpt":"LLM-as-a-Judge has emerged as a popular evaluation strategy, where advanced large language models assess generation results in alignment with human instructions. While these models serve as a promising alternative to human annotators, their reliability in multilingual evaluation remains uncertain. To bridge this gap, we conduct a comprehensive analysis of multilingual LLM-as-a-Judge. Specifically, we evaluate five models from different model families across five diverse tasks involving 25 languages. Our findings reveal that LLMs struggle to achieve consistent judgment results across languages,"},"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":"2505.12201","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-18T02:32:35Z","cross_cats_sorted":[],"title_canon_sha256":"d7a04be750ec2461b026dd0c0f157e711aa32b81cf0ef4db4bf5f04d495a7c74","abstract_canon_sha256":"93759e23e51c821eb9f09a1e359739845ecafcdd2535fe5cae3795a9b45e1f17"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:00.089778Z","signature_b64":"gtx/mRUKbTu6P9TA0XRNP/pIHxg6BM8NCHpWe8B/0giC+XytFDQ6FciVlGa3E9qTzyJyAYiFgAnUFuGBTKhQAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28954afad639a4cd04dd7d665c7d808329cd4796a5d99f8df5a1dc266c27e86d","last_reissued_at":"2026-07-05T11:05:00.089287Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:00.089287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Reliable is Multilingual LLM-as-a-Judge?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Wei Liu, Xiyan Fu","submitted_at":"2025-05-18T02:32:35Z","abstract_excerpt":"LLM-as-a-Judge has emerged as a popular evaluation strategy, where advanced large language models assess generation results in alignment with human instructions. While these models serve as a promising alternative to human annotators, their reliability in multilingual evaluation remains uncertain. To bridge this gap, we conduct a comprehensive analysis of multilingual LLM-as-a-Judge. Specifically, we evaluate five models from different model families across five diverse tasks involving 25 languages. Our findings reveal that LLMs struggle to achieve consistent judgment results across languages,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.12201","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/2505.12201/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":"2505.12201","created_at":"2026-07-05T11:05:00.089361+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.12201v1","created_at":"2026-07-05T11:05:00.089361+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.12201","created_at":"2026-07-05T11:05:00.089361+00:00"},{"alias_kind":"pith_short_12","alias_value":"FCKUV6WWHGSM","created_at":"2026-07-05T11:05:00.089361+00:00"},{"alias_kind":"pith_short_16","alias_value":"FCKUV6WWHGSM2BG5","created_at":"2026-07-05T11:05:00.089361+00:00"},{"alias_kind":"pith_short_8","alias_value":"FCKUV6WW","created_at":"2026-07-05T11:05:00.089361+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03532","citing_title":"Cross-Lingual Bias in Large Language Models: A Comparative Analysis of English and Swahili","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM","json":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM.json","graph_json":"https://pith.science/api/pith-number/FCKUV6WWHGSM2BG5PVTFY7MAQM/graph.json","events_json":"https://pith.science/api/pith-number/FCKUV6WWHGSM2BG5PVTFY7MAQM/events.json","paper":"https://pith.science/paper/FCKUV6WW"},"agent_actions":{"view_html":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM","download_json":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM.json","view_paper":"https://pith.science/paper/FCKUV6WW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.12201&json=true","fetch_graph":"https://pith.science/api/pith-number/FCKUV6WWHGSM2BG5PVTFY7MAQM/graph.json","fetch_events":"https://pith.science/api/pith-number/FCKUV6WWHGSM2BG5PVTFY7MAQM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM/action/storage_attestation","attest_author":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM/action/author_attestation","sign_citation":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM/action/citation_signature","submit_replication":"https://pith.science/pith/FCKUV6WWHGSM2BG5PVTFY7MAQM/action/replication_record"}},"created_at":"2026-07-05T11:05:00.089361+00:00","updated_at":"2026-07-05T11:05:00.089361+00:00"}