{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3YUWILYXT7Q4QB5YCCUFIR52QN","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"1da6ea82542404a748bbeec8cd42f592b95c6ee4ab05852ba71726ab777ab89f","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T15:53:24Z","title_canon_sha256":"7458c02765060a5070340ff3cec80a510cec9c96029809fe4681930330057cac"},"schema_version":"1.0","source":{"id":"2410.08928","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.08928","created_at":"2026-07-05T09:21:49Z"},{"alias_kind":"arxiv_version","alias_value":"2410.08928v2","created_at":"2026-07-05T09:21:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.08928","created_at":"2026-07-05T09:21:49Z"},{"alias_kind":"pith_short_12","alias_value":"3YUWILYXT7Q4","created_at":"2026-07-05T09:21:49Z"},{"alias_kind":"pith_short_16","alias_value":"3YUWILYXT7Q4QB5Y","created_at":"2026-07-05T09:21:49Z"},{"alias_kind":"pith_short_8","alias_value":"3YUWILYX","created_at":"2026-07-05T09:21:49Z"}],"graph_snapshots":[{"event_id":"sha256:a40f5dc0841a65ad653cf84e26605760c868279fae26dc74e6db1d8bce45777d","target":"graph","created_at":"2026-07-05T09:21:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2410.08928/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The rise of Large Language Models (LLMs) has revolutionized natural language processing across numerous languages and tasks. However, evaluating LLM performance in a consistent and meaningful way across multiple European languages remains challenging, especially due to the scarcity of language-parallel multilingual benchmarks. We introduce a multilingual evaluation approach tailored for European languages. We employ translated versions of five widely-used benchmarks to assess the capabilities of 40 LLMs across 21 European languages. Our contributions include examining the effectiveness of tran","authors_text":"Alex Jude, Bernhard Stadler, Fabio Barth, Jasper Schulze Buschhoff, Joachim k\\\"ohler, Johannes Leveling, Klaudia Thellmann, Mehdi Ali, Michael Fromm, Nicolas Flores-Herr, Ren\\'e J\\\"akel","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T15:53:24Z","title":"Towards Multilingual LLM Evaluation for European Languages"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.08928","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e8c9340d229bb4f7cdfe8ccb91b53752203426d9834840649c972641450a2214","target":"record","created_at":"2026-07-05T09:21:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"1da6ea82542404a748bbeec8cd42f592b95c6ee4ab05852ba71726ab777ab89f","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T15:53:24Z","title_canon_sha256":"7458c02765060a5070340ff3cec80a510cec9c96029809fe4681930330057cac"},"schema_version":"1.0","source":{"id":"2410.08928","kind":"arxiv","version":2}},"canonical_sha256":"de29642f179fe1c807b810a85447ba8362313ff0721ba1e6d86e44191e071242","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"de29642f179fe1c807b810a85447ba8362313ff0721ba1e6d86e44191e071242","first_computed_at":"2026-07-05T09:21:49.847661Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:21:49.847661Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mmsFLCe4fIc+cAp/pINCK6XlHtV3E141+3otndfy+2pFWzKrSjjRtGcJIQsNs2FZyFduDapqqWohJI1JTSxkCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:21:49.848114Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.08928","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e8c9340d229bb4f7cdfe8ccb91b53752203426d9834840649c972641450a2214","sha256:a40f5dc0841a65ad653cf84e26605760c868279fae26dc74e6db1d8bce45777d"],"state_sha256":"6ac0dd67bbdc3bba83096d2e183db34f4b1050e6f5ccbc98634b5c4d9c46830d"}