{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BQPOMELTDGS2EM5WHCDJYKJB2G","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":"1c2ce2379c524e528dd5e684ee154b087fe60d368f5cc9a6b29d95b3c43e6874","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-26T16:15:14Z","title_canon_sha256":"fd8cbef4b8b9074d4248da80e5dcacedc7a75341661c44ebeae2fcb59e44b488"},"schema_version":"1.0","source":{"id":"2409.18006","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.18006","created_at":"2026-07-05T09:19:32Z"},{"alias_kind":"arxiv_version","alias_value":"2409.18006v3","created_at":"2026-07-05T09:19:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.18006","created_at":"2026-07-05T09:19:32Z"},{"alias_kind":"pith_short_12","alias_value":"BQPOMELTDGS2","created_at":"2026-07-05T09:19:32Z"},{"alias_kind":"pith_short_16","alias_value":"BQPOMELTDGS2EM5W","created_at":"2026-07-05T09:19:32Z"},{"alias_kind":"pith_short_8","alias_value":"BQPOMELT","created_at":"2026-07-05T09:19:32Z"}],"graph_snapshots":[{"event_id":"sha256:9c743606e2990014067e2c13d3b825907f833a3320ec1dccfbf249a4eaa0becd","target":"graph","created_at":"2026-07-05T09:19:32Z","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/2409.18006/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent large language models (LLMs) demonstrate impressive capabilities in handling long contexts, some exhibiting near-perfect recall on synthetic retrieval tasks. However, these evaluations have mainly focused on English text and involved a single target sentence within lengthy contexts. Our work investigates how LLM performance generalizes to multilingual settings with multiple hidden target sentences. We create a new dataset -- mLongRR -- to comprehensively evaluate several multilingual long-context LLMs on retrieval and reasoning tasks across five languages: English, Vietnamese, Indonesia","authors_text":"Ameeta Agrawal, Andy Dang, Rhitabrat Pokharel, Russell Scheinberg, Sina Bagheri Nezhad","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-26T16:15:14Z","title":"Evaluating Multilingual Long-Context Models for Retrieval and Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.18006","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":"sha256:456468d597fc953274a8f62c50df1c7da3805473acd9095768187ed6c2ee8799","target":"record","created_at":"2026-07-05T09:19:32Z","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":"1c2ce2379c524e528dd5e684ee154b087fe60d368f5cc9a6b29d95b3c43e6874","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-26T16:15:14Z","title_canon_sha256":"fd8cbef4b8b9074d4248da80e5dcacedc7a75341661c44ebeae2fcb59e44b488"},"schema_version":"1.0","source":{"id":"2409.18006","kind":"arxiv","version":3}},"canonical_sha256":"0c1ee6117319a5a233b638869c2921d1ab5e32e8be3a9274ad1c13569878eafa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0c1ee6117319a5a233b638869c2921d1ab5e32e8be3a9274ad1c13569878eafa","first_computed_at":"2026-07-05T09:19:32.149864Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:19:32.149864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"y2vplJRIKOsnRcVPrOF4G5unwjZFoHnxmOgabBH438TYf53dK+vastEVU5E2DqTH4v3ZSjQ95edtjWYps1elAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:19:32.150410Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.18006","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:456468d597fc953274a8f62c50df1c7da3805473acd9095768187ed6c2ee8799","sha256:9c743606e2990014067e2c13d3b825907f833a3320ec1dccfbf249a4eaa0becd"],"state_sha256":"6b42619d64eeaea00ee2499e659e65f12e1e4262b112e949a79fc31d9b3e02e9"}