{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UIAF64RLKTIVSYZMM4JILCMGBC","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":"4a6a1c4e173c0d2f67fe23da0cae3aa8ac1662943b6cea80f008c31d09ae399f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T06:27:04Z","title_canon_sha256":"bd11d2311a464c7923a05aee2856328f107105ebcd6f59ea77a5b9b2240f7087"},"schema_version":"1.0","source":{"id":"2502.04688","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.04688","created_at":"2026-07-05T10:10:55Z"},{"alias_kind":"arxiv_version","alias_value":"2502.04688v1","created_at":"2026-07-05T10:10:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.04688","created_at":"2026-07-05T10:10:55Z"},{"alias_kind":"pith_short_12","alias_value":"UIAF64RLKTIV","created_at":"2026-07-05T10:10:55Z"},{"alias_kind":"pith_short_16","alias_value":"UIAF64RLKTIVSYZM","created_at":"2026-07-05T10:10:55Z"},{"alias_kind":"pith_short_8","alias_value":"UIAF64RL","created_at":"2026-07-05T10:10:55Z"}],"graph_snapshots":[{"event_id":"sha256:e78806155aa044f1df8d8655d1859811ae357d8fa8f2988171df252459b811a1","target":"graph","created_at":"2026-07-05T10:10:55Z","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/2502.04688/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction following is a core capability of modern Large language models (LLMs), making evaluating this capability essential to understanding these models. The Instruction Following Evaluation (IFEval) benchmark from the literature does this using objective criteria, offering a measure of LLM performance without subjective AI or human judgement. However, it only includes English instructions, limiting its ability to assess LLMs in other languages.\n  We propose the Multilingual Instruction Following Evaluation (M-IFEval) benchmark, expanding the evaluation to French, Japanese, and Spanish, wi","authors_text":"Andrea Carde\\~na D\\'iaz, Antoine Dussolle, Peter Devine, Shota Sato","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T06:27:04Z","title":"M-IFEval: Multilingual Instruction-Following Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.04688","kind":"arxiv","version":1},"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:ff4fb06b7c2e7619bf95e22240b09201e60e8342d07805eb5862c7c7fb92f242","target":"record","created_at":"2026-07-05T10:10:55Z","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":"4a6a1c4e173c0d2f67fe23da0cae3aa8ac1662943b6cea80f008c31d09ae399f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T06:27:04Z","title_canon_sha256":"bd11d2311a464c7923a05aee2856328f107105ebcd6f59ea77a5b9b2240f7087"},"schema_version":"1.0","source":{"id":"2502.04688","kind":"arxiv","version":1}},"canonical_sha256":"a2005f722b54d159632c671285898608ae23c2c4ca49a5f0e9e5a22cdbbb45a9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a2005f722b54d159632c671285898608ae23c2c4ca49a5f0e9e5a22cdbbb45a9","first_computed_at":"2026-07-05T10:10:55.266406Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:10:55.266406Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Xmlvac8mrm4/9jUtfgneF6IKLhD0rCzdq3Uh58UTsOo2jFDxKp3G9VU7EUBpARnMwWFTej4Oa3uhy+wUtFRhDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:10:55.266819Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.04688","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ff4fb06b7c2e7619bf95e22240b09201e60e8342d07805eb5862c7c7fb92f242","sha256:e78806155aa044f1df8d8655d1859811ae357d8fa8f2988171df252459b811a1"],"state_sha256":"6c8e4cf99855df28841037a612c8636257c8f8b82924cb362b2b7b472f6b5bf4"}