{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:75UWU7L3ZDF3NPN6RO5AHSS2LE","short_pith_number":"pith:75UWU7L3","canonical_record":{"source":{"id":"2501.14851","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T15:49:10Z","cross_cats_sorted":["cs.AI","cs.LG","cs.LO"],"title_canon_sha256":"55f827021fe71271e51f874e169ba031f03f158e4cc261031958bafb4a49e91a","abstract_canon_sha256":"da2204811f7050b675037da6afeb40a47315ff387ffae9dc6f71fc1ff3d8ef88"},"schema_version":"1.0"},"canonical_sha256":"ff696a7d7bc8cbb6bdbe8bba03ca5a593237b61b2c27b7deae5dbc0f107cf0f1","source":{"kind":"arxiv","id":"2501.14851","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.14851","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"arxiv_version","alias_value":"2501.14851v2","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.14851","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"pith_short_12","alias_value":"75UWU7L3ZDF3","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"pith_short_16","alias_value":"75UWU7L3ZDF3NPN6","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"pith_short_8","alias_value":"75UWU7L3","created_at":"2026-07-05T11:00:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:75UWU7L3ZDF3NPN6RO5AHSS2LE","target":"record","payload":{"canonical_record":{"source":{"id":"2501.14851","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T15:49:10Z","cross_cats_sorted":["cs.AI","cs.LG","cs.LO"],"title_canon_sha256":"55f827021fe71271e51f874e169ba031f03f158e4cc261031958bafb4a49e91a","abstract_canon_sha256":"da2204811f7050b675037da6afeb40a47315ff387ffae9dc6f71fc1ff3d8ef88"},"schema_version":"1.0"},"canonical_sha256":"ff696a7d7bc8cbb6bdbe8bba03ca5a593237b61b2c27b7deae5dbc0f107cf0f1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:41.801851Z","signature_b64":"A+21YJbWAf15JVkU9fm1NSh48PawbchMdXyNXNPxlYHioWfFxAdBDbxHA7LbuJtVLO6NHBUWpCFS3wxutzFyAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff696a7d7bc8cbb6bdbe8bba03ca5a593237b61b2c27b7deae5dbc0f107cf0f1","last_reissued_at":"2026-07-05T11:00:41.801311Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:41.801311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.14851","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":"2026-07-05T11:00:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ndj9vSYplfmYLmNTNwmH9VX6h2LQ+jgnBtFQmpx9ljKEfylsUbF2mNENg+U3KPzwwC06F1kZHgoB7RcHJXq9Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T11:37:12.221434Z"},"content_sha256":"ed61f126fc5a01d1c1179d24dd7f6fa5c94024dae924d393cfb628e43affb7cf","schema_version":"1.0","event_id":"sha256:ed61f126fc5a01d1c1179d24dd7f6fa5c94024dae924d393cfb628e43affb7cf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:75UWU7L3ZDF3NPN6RO5AHSS2LE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"JustLogic: A Comprehensive Benchmark for Evaluating Deductive Reasoning in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.LO"],"primary_cat":"cs.CL","authors_text":"Dacheng Tao, Michael K. Chen, Xikun Zhang","submitted_at":"2025-01-24T15:49:10Z","abstract_excerpt":"Logical reasoning is a critical component of Large Language Models (LLMs), and substantial research efforts in recent years have aimed to enhance their deductive reasoning capabilities. However, existing deductive reasoning benchmarks, which are crucial for evaluating and advancing LLMs, are inadequate due to their lack of task complexity, presence of prior knowledge as a confounder, and superficial error analysis. To address these deficiencies, we introduce JustLogic, a synthetically generated deductive reasoning benchmark designed for rigorous evaluation of LLMs. JustLogic is (i) highly comp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.14851","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/2501.14851/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":"2026-07-05T11:00:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2ksnwCq3hHfYDGFRcG+aPL5IMQ9UgwzQFKCoUfVpBr9SZo7lt+llGPOmG0Hqzz/3z73VDWVcpVI1bMlEjq9rDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T11:37:12.221931Z"},"content_sha256":"6487b1e737e777d61e4da3fe374726ba293464ac802a1f09c977e3b40f89bc58","schema_version":"1.0","event_id":"sha256:6487b1e737e777d61e4da3fe374726ba293464ac802a1f09c977e3b40f89bc58"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE/bundle.json","state_url":"https://pith.science/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE/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":"2026-08-07T11:37:12Z","links":{"resolver":"https://pith.science/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE","bundle":"https://pith.science/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE/bundle.json","state":"https://pith.science/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/75UWU7L3ZDF3NPN6RO5AHSS2LE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:75UWU7L3ZDF3NPN6RO5AHSS2LE","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":"da2204811f7050b675037da6afeb40a47315ff387ffae9dc6f71fc1ff3d8ef88","cross_cats_sorted":["cs.AI","cs.LG","cs.LO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T15:49:10Z","title_canon_sha256":"55f827021fe71271e51f874e169ba031f03f158e4cc261031958bafb4a49e91a"},"schema_version":"1.0","source":{"id":"2501.14851","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.14851","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"arxiv_version","alias_value":"2501.14851v2","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.14851","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"pith_short_12","alias_value":"75UWU7L3ZDF3","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"pith_short_16","alias_value":"75UWU7L3ZDF3NPN6","created_at":"2026-07-05T11:00:41Z"},{"alias_kind":"pith_short_8","alias_value":"75UWU7L3","created_at":"2026-07-05T11:00:41Z"}],"graph_snapshots":[{"event_id":"sha256:6487b1e737e777d61e4da3fe374726ba293464ac802a1f09c977e3b40f89bc58","target":"graph","created_at":"2026-07-05T11:00:41Z","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/2501.14851/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Logical reasoning is a critical component of Large Language Models (LLMs), and substantial research efforts in recent years have aimed to enhance their deductive reasoning capabilities. However, existing deductive reasoning benchmarks, which are crucial for evaluating and advancing LLMs, are inadequate due to their lack of task complexity, presence of prior knowledge as a confounder, and superficial error analysis. To address these deficiencies, we introduce JustLogic, a synthetically generated deductive reasoning benchmark designed for rigorous evaluation of LLMs. JustLogic is (i) highly comp","authors_text":"Dacheng Tao, Michael K. Chen, Xikun Zhang","cross_cats":["cs.AI","cs.LG","cs.LO"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T15:49:10Z","title":"JustLogic: A Comprehensive Benchmark for Evaluating Deductive Reasoning in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.14851","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:ed61f126fc5a01d1c1179d24dd7f6fa5c94024dae924d393cfb628e43affb7cf","target":"record","created_at":"2026-07-05T11:00:41Z","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":"da2204811f7050b675037da6afeb40a47315ff387ffae9dc6f71fc1ff3d8ef88","cross_cats_sorted":["cs.AI","cs.LG","cs.LO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T15:49:10Z","title_canon_sha256":"55f827021fe71271e51f874e169ba031f03f158e4cc261031958bafb4a49e91a"},"schema_version":"1.0","source":{"id":"2501.14851","kind":"arxiv","version":2}},"canonical_sha256":"ff696a7d7bc8cbb6bdbe8bba03ca5a593237b61b2c27b7deae5dbc0f107cf0f1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ff696a7d7bc8cbb6bdbe8bba03ca5a593237b61b2c27b7deae5dbc0f107cf0f1","first_computed_at":"2026-07-05T11:00:41.801311Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:00:41.801311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"A+21YJbWAf15JVkU9fm1NSh48PawbchMdXyNXNPxlYHioWfFxAdBDbxHA7LbuJtVLO6NHBUWpCFS3wxutzFyAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:00:41.801851Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.14851","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ed61f126fc5a01d1c1179d24dd7f6fa5c94024dae924d393cfb628e43affb7cf","sha256:6487b1e737e777d61e4da3fe374726ba293464ac802a1f09c977e3b40f89bc58"],"state_sha256":"a55207d1b3a51c1ce5b7b32d4017876f99d5f444fbd3a4b3b3f43dd39399900e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MZOq+EXSMvNNBztfC9BUrTOQvEc+OG1u5DyR4LyRWnnGe164nJG170tqMcEjwNjsf1IB5N1q14RQF2bqn+krBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T11:37:12.226579Z","bundle_sha256":"1a7ffcb3f472b0451ce68bc28f1e05f33a9d9e28f35e8f7bc59f8855984f21c6"}}