{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:OBOD24ZHYNF3KQFYLIT4TYHXA7","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":"c0c30f83c65c5bfca3f9a859d70df5aef5c18f5efaa6ad0e980ec169d6d328cd","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T16:35:30Z","title_canon_sha256":"f9dd3449b13ed1c1f78fd9e6387d75f6d294e505f8a23e6340c39fd93796747a"},"schema_version":"1.0","source":{"id":"2405.06680","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.06680","created_at":"2026-07-05T09:18:25Z"},{"alias_kind":"arxiv_version","alias_value":"2405.06680v4","created_at":"2026-07-05T09:18:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06680","created_at":"2026-07-05T09:18:25Z"},{"alias_kind":"pith_short_12","alias_value":"OBOD24ZHYNF3","created_at":"2026-07-05T09:18:25Z"},{"alias_kind":"pith_short_16","alias_value":"OBOD24ZHYNF3KQFY","created_at":"2026-07-05T09:18:25Z"},{"alias_kind":"pith_short_8","alias_value":"OBOD24ZH","created_at":"2026-07-05T09:18:25Z"}],"graph_snapshots":[{"event_id":"sha256:372d826544e3682b0408576f283fc3a5a5c0f3503c2eaf6e368561f2d0bf5617","target":"graph","created_at":"2026-07-05T09:18:25Z","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/2405.06680/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human cognition exhibits systematic compositionality, the algebraic ability to generate infinite novel combinations from finite learned components, which is the key to understanding and reasoning about complex logic. In this work, we investigate the compositionality of large language models (LLMs) in mathematical reasoning. Specifically, we construct a new dataset \\textsc{MathTrap} by introducing carefully designed logical traps into the problem descriptions of MATH and GSM8K. Since problems with logical flaws are quite rare in the real world, these represent \"unseen\" cases to LLMs. Solving th","authors_text":"Jingqi Tong, Jun Zhao, Ming Zhang, Qi Zhang, Xuanjing Huang, Yurong Mou","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T16:35:30Z","title":"Exploring the Compositional Deficiency of Large Language Models in Mathematical Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06680","kind":"arxiv","version":4},"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":"sha256:0832361b3d5b109eef87449e32efb67518e1e217b47c796f1e131a0397242b59","target":"record","created_at":"2026-07-05T09:18:25Z","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":"c0c30f83c65c5bfca3f9a859d70df5aef5c18f5efaa6ad0e980ec169d6d328cd","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T16:35:30Z","title_canon_sha256":"f9dd3449b13ed1c1f78fd9e6387d75f6d294e505f8a23e6340c39fd93796747a"},"schema_version":"1.0","source":{"id":"2405.06680","kind":"arxiv","version":4}},"canonical_sha256":"705c3d7327c34bb540b85a27c9e0f707dfb2fa221dde30bd388e8f84817876e8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"705c3d7327c34bb540b85a27c9e0f707dfb2fa221dde30bd388e8f84817876e8","first_computed_at":"2026-07-05T09:18:25.517691Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:18:25.517691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3IZU9kHY5SWDBLyWLSdEu8wkTf5gh/srZCwZUxOI2fE5Mvn96uNxWUZtzY3tY+KI9KEAraEvhkpH83ghSlqdBA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:18:25.518180Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.06680","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0832361b3d5b109eef87449e32efb67518e1e217b47c796f1e131a0397242b59","sha256:372d826544e3682b0408576f283fc3a5a5c0f3503c2eaf6e368561f2d0bf5617"],"state_sha256":"f85c4b7ad66a468a175036da6aa3175b3f1722fa676090d7c4567151305fd260"}