{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:WH2RK4HEQST323WBEU5IMKP3GA","short_pith_number":"pith:WH2RK4HE","canonical_record":{"source":{"id":"2412.13377","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T23:25:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b718e8a0570baee8335a42e64a00ffcb45c8cfa01091c4cbec18c1836df80f7d","abstract_canon_sha256":"40deb6f77a7cb179d870d2868f6ab1d0505d3b2ab4b4bcee672acdbf313118ae"},"schema_version":"1.0"},"canonical_sha256":"b1f51570e484a7bd6ec1253a8629fb302d1e577bc4539687ceec48f2606b22fa","source":{"kind":"arxiv","id":"2412.13377","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.13377","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"arxiv_version","alias_value":"2412.13377v2","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.13377","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"pith_short_12","alias_value":"WH2RK4HEQST3","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"pith_short_16","alias_value":"WH2RK4HEQST323WB","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"pith_short_8","alias_value":"WH2RK4HE","created_at":"2026-07-05T11:04:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:WH2RK4HEQST323WBEU5IMKP3GA","target":"record","payload":{"canonical_record":{"source":{"id":"2412.13377","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T23:25:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b718e8a0570baee8335a42e64a00ffcb45c8cfa01091c4cbec18c1836df80f7d","abstract_canon_sha256":"40deb6f77a7cb179d870d2868f6ab1d0505d3b2ab4b4bcee672acdbf313118ae"},"schema_version":"1.0"},"canonical_sha256":"b1f51570e484a7bd6ec1253a8629fb302d1e577bc4539687ceec48f2606b22fa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:42.914340Z","signature_b64":"sLNXWBmerlBoIwutUKpvkJSW3vVPaw4gsh8TdrgPJkJsPSi11gaNSQAXmE3N3P94AoSY2Qg0EEMvTHazgZo+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b1f51570e484a7bd6ec1253a8629fb302d1e577bc4539687ceec48f2606b22fa","last_reissued_at":"2026-07-05T11:04:42.913911Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:42.913911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.13377","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:04:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XxhHlPNGF8hRVQk9KVZv96BWstYHF+mZPvuNOwofcZdmNPgbPqIbNgonnrrxR7PYzkAWSZ1pArCjwo7KINzhCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T14:25:04.077446Z"},"content_sha256":"b7d1b847c8175201bec3cfa12254ace21241eb302a3d45e15e797125291749c1","schema_version":"1.0","event_id":"sha256:b7d1b847c8175201bec3cfa12254ace21241eb302a3d45e15e797125291749c1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:WH2RK4HEQST323WBEU5IMKP3GA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DateLogicQA: Benchmarking Temporal Biases in Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cristina Mahanta, Gagan Bhatia, Madiha Kazi, Mingze Tang","submitted_at":"2024-12-17T23:25:47Z","abstract_excerpt":"This paper introduces DateLogicQA, a benchmark with 190 questions covering diverse date formats, temporal contexts, and reasoning types. We propose the Semantic Integrity Metric to assess tokenization quality and analyse two biases: Representation-Level Bias, affecting embeddings, and Logical-Level Bias, influencing reasoning outputs. Our findings provide a comprehensive evaluation of LLMs' capabilities and limitations in temporal reasoning, highlighting key challenges in handling temporal data accurately."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.13377","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2412.13377/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:04:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hWhZAuFMAQE0DWwkeOLROEetUODVLD+hcu707iZSg9MltQVe0b8pQ1qZohsX7gH4qxJCA6q6CEfRGNtKtCk/Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T14:25:04.077992Z"},"content_sha256":"5f536a1aad2d17071ecc9effed3e1bf6beb68d074b991da3b3d40fb3a588d9f9","schema_version":"1.0","event_id":"sha256:5f536a1aad2d17071ecc9effed3e1bf6beb68d074b991da3b3d40fb3a588d9f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WH2RK4HEQST323WBEU5IMKP3GA/bundle.json","state_url":"https://pith.science/pith/WH2RK4HEQST323WBEU5IMKP3GA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WH2RK4HEQST323WBEU5IMKP3GA/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-14T14:25:04Z","links":{"resolver":"https://pith.science/pith/WH2RK4HEQST323WBEU5IMKP3GA","bundle":"https://pith.science/pith/WH2RK4HEQST323WBEU5IMKP3GA/bundle.json","state":"https://pith.science/pith/WH2RK4HEQST323WBEU5IMKP3GA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WH2RK4HEQST323WBEU5IMKP3GA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:WH2RK4HEQST323WBEU5IMKP3GA","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":"40deb6f77a7cb179d870d2868f6ab1d0505d3b2ab4b4bcee672acdbf313118ae","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T23:25:47Z","title_canon_sha256":"b718e8a0570baee8335a42e64a00ffcb45c8cfa01091c4cbec18c1836df80f7d"},"schema_version":"1.0","source":{"id":"2412.13377","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.13377","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"arxiv_version","alias_value":"2412.13377v2","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.13377","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"pith_short_12","alias_value":"WH2RK4HEQST3","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"pith_short_16","alias_value":"WH2RK4HEQST323WB","created_at":"2026-07-05T11:04:42Z"},{"alias_kind":"pith_short_8","alias_value":"WH2RK4HE","created_at":"2026-07-05T11:04:42Z"}],"graph_snapshots":[{"event_id":"sha256:5f536a1aad2d17071ecc9effed3e1bf6beb68d074b991da3b3d40fb3a588d9f9","target":"graph","created_at":"2026-07-05T11:04:42Z","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/2412.13377/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces DateLogicQA, a benchmark with 190 questions covering diverse date formats, temporal contexts, and reasoning types. We propose the Semantic Integrity Metric to assess tokenization quality and analyse two biases: Representation-Level Bias, affecting embeddings, and Logical-Level Bias, influencing reasoning outputs. Our findings provide a comprehensive evaluation of LLMs' capabilities and limitations in temporal reasoning, highlighting key challenges in handling temporal data accurately.","authors_text":"Cristina Mahanta, Gagan Bhatia, Madiha Kazi, Mingze Tang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T23:25:47Z","title":"DateLogicQA: Benchmarking Temporal Biases in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.13377","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:b7d1b847c8175201bec3cfa12254ace21241eb302a3d45e15e797125291749c1","target":"record","created_at":"2026-07-05T11:04:42Z","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":"40deb6f77a7cb179d870d2868f6ab1d0505d3b2ab4b4bcee672acdbf313118ae","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T23:25:47Z","title_canon_sha256":"b718e8a0570baee8335a42e64a00ffcb45c8cfa01091c4cbec18c1836df80f7d"},"schema_version":"1.0","source":{"id":"2412.13377","kind":"arxiv","version":2}},"canonical_sha256":"b1f51570e484a7bd6ec1253a8629fb302d1e577bc4539687ceec48f2606b22fa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b1f51570e484a7bd6ec1253a8629fb302d1e577bc4539687ceec48f2606b22fa","first_computed_at":"2026-07-05T11:04:42.913911Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:04:42.913911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sLNXWBmerlBoIwutUKpvkJSW3vVPaw4gsh8TdrgPJkJsPSi11gaNSQAXmE3N3P94AoSY2Qg0EEMvTHazgZo+DA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:04:42.914340Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.13377","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b7d1b847c8175201bec3cfa12254ace21241eb302a3d45e15e797125291749c1","sha256:5f536a1aad2d17071ecc9effed3e1bf6beb68d074b991da3b3d40fb3a588d9f9"],"state_sha256":"5f4fdba6afb47f53d04c3a4841d9f910a804fc75a6ff42c3acd851a865135851"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kwxsXmItLNZ3sjVvZOROi2VoQ29LA2iLsYJxS3gsvAInpMdUfG17V8Evr2y7ZXTLcM5fO1pLw/xC/hhApyAPAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T14:25:04.084059Z","bundle_sha256":"25dc6d52bad33cd5513b90a6fa234079d93d9ad8d3818df1d67391f2688a5adf"}}