{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OALFNBDG4RJL6AJWIAZQWTMIP4","short_pith_number":"pith:OALFNBDG","schema_version":"1.0","canonical_sha256":"7016568466e452bf013640330b4d887f30f81b1da4758911c44b68f9bd301e0a","source":{"kind":"arxiv","id":"2402.03877","version":3},"attestation_state":"computed","paper":{"title":"Beyond Lines and Circles: Unveiling the Geometric Reasoning Gap in Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Henryk Michalewski, Mateusz Malinowski, Spyridon Mouselinos","submitted_at":"2024-02-06T10:37:21Z","abstract_excerpt":"Large Language Models (LLMs) demonstrate ever-increasing abilities in mathematical and algorithmic tasks, yet their geometric reasoning skills are underexplored. We investigate LLMs' abilities in constructive geometric problem-solving one of the most fundamental steps in the development of human mathematical reasoning. Our work reveals notable challenges that the state-of-the-art LLMs face in this domain despite many successes in similar areas. LLMs exhibit biases in target variable selection and struggle with 2D spatial relationships, often misrepresenting and hallucinating objects and their "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2402.03877","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-06T10:37:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8db9275da406b8819f6965c3fd30a1b10ed5479c08c551e60cf5f7fe31983e56","abstract_canon_sha256":"9fdc33a3fcdddfc47041da19e6e65f8fbd5ca7aa5be9edde73bc1a98b8b213fd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:09:19.426636Z","signature_b64":"l2UISXmRhAOd7C83nssMZDHGVFQ5Mq03c4jHnlhSs8hfXkiCybMvSasQGssB/ratb7LnO2RUNTViXaHyeON8Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7016568466e452bf013640330b4d887f30f81b1da4758911c44b68f9bd301e0a","last_reissued_at":"2026-07-05T09:09:19.426116Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:09:19.426116Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Lines and Circles: Unveiling the Geometric Reasoning Gap in Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Henryk Michalewski, Mateusz Malinowski, Spyridon Mouselinos","submitted_at":"2024-02-06T10:37:21Z","abstract_excerpt":"Large Language Models (LLMs) demonstrate ever-increasing abilities in mathematical and algorithmic tasks, yet their geometric reasoning skills are underexplored. We investigate LLMs' abilities in constructive geometric problem-solving one of the most fundamental steps in the development of human mathematical reasoning. Our work reveals notable challenges that the state-of-the-art LLMs face in this domain despite many successes in similar areas. LLMs exhibit biases in target variable selection and struggle with 2D spatial relationships, often misrepresenting and hallucinating objects and their "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.03877","kind":"arxiv","version":3},"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/2402.03877/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2402.03877","created_at":"2026-07-05T09:09:19.426181+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.03877v3","created_at":"2026-07-05T09:09:19.426181+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.03877","created_at":"2026-07-05T09:09:19.426181+00:00"},{"alias_kind":"pith_short_12","alias_value":"OALFNBDG4RJL","created_at":"2026-07-05T09:09:19.426181+00:00"},{"alias_kind":"pith_short_16","alias_value":"OALFNBDG4RJL6AJW","created_at":"2026-07-05T09:09:19.426181+00:00"},{"alias_kind":"pith_short_8","alias_value":"OALFNBDG","created_at":"2026-07-05T09:09:19.426181+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.02871","citing_title":"Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoning","ref_index":134,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4","json":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4.json","graph_json":"https://pith.science/api/pith-number/OALFNBDG4RJL6AJWIAZQWTMIP4/graph.json","events_json":"https://pith.science/api/pith-number/OALFNBDG4RJL6AJWIAZQWTMIP4/events.json","paper":"https://pith.science/paper/OALFNBDG"},"agent_actions":{"view_html":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4","download_json":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4.json","view_paper":"https://pith.science/paper/OALFNBDG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.03877&json=true","fetch_graph":"https://pith.science/api/pith-number/OALFNBDG4RJL6AJWIAZQWTMIP4/graph.json","fetch_events":"https://pith.science/api/pith-number/OALFNBDG4RJL6AJWIAZQWTMIP4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4/action/storage_attestation","attest_author":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4/action/author_attestation","sign_citation":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4/action/citation_signature","submit_replication":"https://pith.science/pith/OALFNBDG4RJL6AJWIAZQWTMIP4/action/replication_record"}},"created_at":"2026-07-05T09:09:19.426181+00:00","updated_at":"2026-07-05T09:09:19.426181+00:00"}