{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UUSOHELHCFWS4KHOOHCFARJ45C","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":"76a88ef8d181cf5775fb158b81cd517467b4a5c4500c84e4536b95a4574dd8c5","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T02:03:57Z","title_canon_sha256":"7010a90729c271fa3ac17946c52cd8fc4669f722a8ec035236cc17cfe5a86017"},"schema_version":"1.0","source":{"id":"2406.16264","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.16264","created_at":"2026-07-05T09:23:52Z"},{"alias_kind":"arxiv_version","alias_value":"2406.16264v3","created_at":"2026-07-05T09:23:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.16264","created_at":"2026-07-05T09:23:52Z"},{"alias_kind":"pith_short_12","alias_value":"UUSOHELHCFWS","created_at":"2026-07-05T09:23:52Z"},{"alias_kind":"pith_short_16","alias_value":"UUSOHELHCFWS4KHO","created_at":"2026-07-05T09:23:52Z"},{"alias_kind":"pith_short_8","alias_value":"UUSOHELH","created_at":"2026-07-05T09:23:52Z"}],"graph_snapshots":[{"event_id":"sha256:75153f49552fd3ec5e5c982dd65f82a9d0379aab2bddee95abcd85b1f251f033","target":"graph","created_at":"2026-07-05T09:23:52Z","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/2406.16264/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Synthetic long-context LLM benchmarks (e.g., \"needle-in-the-haystack\") test only surface-level retrieval capabilities, but how well can long-context LLMs retrieve, synthesize, and reason over information across book-length inputs? We address this question by creating NoCha, a dataset of 1,001 minimally different pairs of true and false claims about 67 recently-published English fictional books, written by human readers of those books. In contrast to existing long-context benchmarks, our annotators confirm that the largest share of pairs in NoCha require global reasoning over the entire book to","authors_text":"Katherine Thai, Kyle Lo, Marzena Karpinska, Mohit Iyyer, Tanya Goyal","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T02:03:57Z","title":"One Thousand and One Pairs: A \"novel\" challenge for long-context language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.16264","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:f03b28775c73e8de9538d70f32c5a07d66799d6fc955ecc292da0d07cb54d1a9","target":"record","created_at":"2026-07-05T09:23:52Z","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":"76a88ef8d181cf5775fb158b81cd517467b4a5c4500c84e4536b95a4574dd8c5","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T02:03:57Z","title_canon_sha256":"7010a90729c271fa3ac17946c52cd8fc4669f722a8ec035236cc17cfe5a86017"},"schema_version":"1.0","source":{"id":"2406.16264","kind":"arxiv","version":3}},"canonical_sha256":"a524e39167116d2e28ee71c450453ce8a9e35806be1e363cd161b3d279c82e2d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a524e39167116d2e28ee71c450453ce8a9e35806be1e363cd161b3d279c82e2d","first_computed_at":"2026-07-05T09:23:52.952963Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:23:52.952963Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"b8pHrCP0tXVwp/hr4S1nl8W+vqPAIwFq+mDIKEAQLzqazubiXfl54ngdOHKs3z6kO2cxjTi13f1hNshJZpxsCg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:23:52.953474Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.16264","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f03b28775c73e8de9538d70f32c5a07d66799d6fc955ecc292da0d07cb54d1a9","sha256:75153f49552fd3ec5e5c982dd65f82a9d0379aab2bddee95abcd85b1f251f033"],"state_sha256":"9caf12f04f6448281b6f8dc7ba666bcecfe1c169f2d42836424f0c7afffde2e9"}