{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DR3QNHMJLWWGL5WKPJP2FCHQGI","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":"e0e63cb4035e71bbaa5dc171699e15e8d9863bc836601ef8a645a02313471c43","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-28T17:03:51Z","title_canon_sha256":"62bc991d9352f9c5ba226f75663340bb0ab422a913e3fee339d651b4a076270b"},"schema_version":"1.0","source":{"id":"2406.20052","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.20052","created_at":"2026-07-05T10:44:14Z"},{"alias_kind":"arxiv_version","alias_value":"2406.20052v3","created_at":"2026-07-05T10:44:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.20052","created_at":"2026-07-05T10:44:14Z"},{"alias_kind":"pith_short_12","alias_value":"DR3QNHMJLWWG","created_at":"2026-07-05T10:44:14Z"},{"alias_kind":"pith_short_16","alias_value":"DR3QNHMJLWWGL5WK","created_at":"2026-07-05T10:44:14Z"},{"alias_kind":"pith_short_8","alias_value":"DR3QNHMJ","created_at":"2026-07-05T10:44:14Z"}],"graph_snapshots":[{"event_id":"sha256:52c5f3261f2d1ed95922e4c84fbd3bbf4234193a8848a6bf289883e71a109a02","target":"graph","created_at":"2026-07-05T10:44:14Z","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/2406.20052/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We investigate a surprising limitation of LLMs: their inability to consistently generate text in a user's desired language. We create the Language Confusion Benchmark (LCB) to evaluate such failures, covering 15 typologically diverse languages with existing and newly-created English and multilingual prompts. We evaluate a range of LLMs on monolingual and cross-lingual generation reflecting practical use cases, finding that Llama Instruct and Mistral models exhibit high degrees of language confusion and even the strongest models fail to consistently respond in the correct language. We observe t","authors_text":"Alexandre B\\'erard, Kelly Marchisio, Sebastian Ruder, Th\\'eo Dehaze, Wei-Yin Ko","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-28T17:03:51Z","title":"Understanding and Mitigating Language Confusion in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.20052","kind":"arxiv","version":3},"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:db51f5391c6cca5c7f24e206185e8c4e2afca393b2c8ba9f6cbfc33539b818c7","target":"record","created_at":"2026-07-05T10:44:14Z","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":"e0e63cb4035e71bbaa5dc171699e15e8d9863bc836601ef8a645a02313471c43","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-28T17:03:51Z","title_canon_sha256":"62bc991d9352f9c5ba226f75663340bb0ab422a913e3fee339d651b4a076270b"},"schema_version":"1.0","source":{"id":"2406.20052","kind":"arxiv","version":3}},"canonical_sha256":"1c77069d895dac65f6ca7a5fa288f03210e436a90b3b113e3bf188460c81db13","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1c77069d895dac65f6ca7a5fa288f03210e436a90b3b113e3bf188460c81db13","first_computed_at":"2026-07-05T10:44:14.989243Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:44:14.989243Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"s1sYGSm2tCd6ikNse5hC+XjxffCuEPyzR5QR8M3BsAI0KeqaVvo3uawbogRFjnEtU79I+043nMi5eOhXOsifCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:44:14.989725Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.20052","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:db51f5391c6cca5c7f24e206185e8c4e2afca393b2c8ba9f6cbfc33539b818c7","sha256:52c5f3261f2d1ed95922e4c84fbd3bbf4234193a8848a6bf289883e71a109a02"],"state_sha256":"5e32f8b1d83350734c06c577c61d3b9fe0512b9353461f19d45db7159603f3b3"}