{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:CITWRTYJDR3PZJACEEY4MAAE2Y","short_pith_number":"pith:CITWRTYJ","schema_version":"1.0","canonical_sha256":"122768cf091c76fca4022131c60004d611be2b26e688bf21e584ef8ff70a735f","source":{"kind":"arxiv","id":"2305.13675","version":2},"attestation_state":"computed","paper":{"title":"Polyglot or Not? Measuring Multilingual Encyclopedic Knowledge in Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Daniel Furman, Shreshta Bhat, Tim Schott","submitted_at":"2023-05-23T04:31:39Z","abstract_excerpt":"In this work, we assess the ability of foundation models to recall encyclopedic knowledge across a wide range of linguistic contexts. To support this, we: 1) produce a 20-language dataset that contains 303k factual associations paired with counterfactuals, 2) evaluate 5 models in a multilingual test, and 3) benchmark a diverse set of 24 models in an English-only test. Meta's LLaMA achieves the highest scores in both multilingual and English-only evaluations. Yet, an analysis of LLaMA's errors reveals significant limitations in its ability to recall facts in languages other than English, plus d"},"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":"2305.13675","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T04:31:39Z","cross_cats_sorted":[],"title_canon_sha256":"3577c32e5a65c25d9f23474f21eaf02ad1b49ce8bd591ff7448f85e5dc31c0d2","abstract_canon_sha256":"ab776bc1d7db0151c05ecb40be41a0713a90d1419a6b355331d87fbdeaaf7c50"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:20:51.064530Z","signature_b64":"YQTyhNR91j72TxUIX7WgMPYzUDsYdqYMSNOBGYb6HeXO9sE8Q3VUVcwFq0aJCk1ahjMjVrzkkERHZVucy4grCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"122768cf091c76fca4022131c60004d611be2b26e688bf21e584ef8ff70a735f","last_reissued_at":"2026-07-05T07:20:51.064104Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:20:51.064104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Polyglot or Not? Measuring Multilingual Encyclopedic Knowledge in Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Daniel Furman, Shreshta Bhat, Tim Schott","submitted_at":"2023-05-23T04:31:39Z","abstract_excerpt":"In this work, we assess the ability of foundation models to recall encyclopedic knowledge across a wide range of linguistic contexts. To support this, we: 1) produce a 20-language dataset that contains 303k factual associations paired with counterfactuals, 2) evaluate 5 models in a multilingual test, and 3) benchmark a diverse set of 24 models in an English-only test. Meta's LLaMA achieves the highest scores in both multilingual and English-only evaluations. Yet, an analysis of LLaMA's errors reveals significant limitations in its ability to recall facts in languages other than English, plus d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13675","kind":"arxiv","version":2},"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/2305.13675/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":"2305.13675","created_at":"2026-07-05T07:20:51.064159+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.13675v2","created_at":"2026-07-05T07:20:51.064159+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13675","created_at":"2026-07-05T07:20:51.064159+00:00"},{"alias_kind":"pith_short_12","alias_value":"CITWRTYJDR3P","created_at":"2026-07-05T07:20:51.064159+00:00"},{"alias_kind":"pith_short_16","alias_value":"CITWRTYJDR3PZJAC","created_at":"2026-07-05T07:20:51.064159+00:00"},{"alias_kind":"pith_short_8","alias_value":"CITWRTYJ","created_at":"2026-07-05T07:20:51.064159+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.02173","citing_title":"Mass-Editing Memory with Attention in Transformers: A cross-lingual exploration of knowledge","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y","json":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y.json","graph_json":"https://pith.science/api/pith-number/CITWRTYJDR3PZJACEEY4MAAE2Y/graph.json","events_json":"https://pith.science/api/pith-number/CITWRTYJDR3PZJACEEY4MAAE2Y/events.json","paper":"https://pith.science/paper/CITWRTYJ"},"agent_actions":{"view_html":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y","download_json":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y.json","view_paper":"https://pith.science/paper/CITWRTYJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.13675&json=true","fetch_graph":"https://pith.science/api/pith-number/CITWRTYJDR3PZJACEEY4MAAE2Y/graph.json","fetch_events":"https://pith.science/api/pith-number/CITWRTYJDR3PZJACEEY4MAAE2Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y/action/storage_attestation","attest_author":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y/action/author_attestation","sign_citation":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y/action/citation_signature","submit_replication":"https://pith.science/pith/CITWRTYJDR3PZJACEEY4MAAE2Y/action/replication_record"}},"created_at":"2026-07-05T07:20:51.064159+00:00","updated_at":"2026-07-05T07:20:51.064159+00:00"}