{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YLXYNEMZUAFOBYXR75FCI3XLKJ","short_pith_number":"pith:YLXYNEMZ","canonical_record":{"source":{"id":"2507.07313","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-09T22:22:49Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9ab6d82c54367a6f3f9080bcaed81477d3ff07dd06e710ff0ff514e798992138","abstract_canon_sha256":"d4caac9086db6ad8d66e0b03e54246fe861ed9f41b630a41b729f8a86cfadd56"},"schema_version":"1.0"},"canonical_sha256":"c2ef869199a00ae0e2f1ff4a246eeb526f451c1ecb08c66225062df19ce94f75","source":{"kind":"arxiv","id":"2507.07313","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07313","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07313v1","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07313","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_12","alias_value":"YLXYNEMZUAFO","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_16","alias_value":"YLXYNEMZUAFOBYXR","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_8","alias_value":"YLXYNEMZ","created_at":"2026-07-05T11:34:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YLXYNEMZUAFOBYXR75FCI3XLKJ","target":"record","payload":{"canonical_record":{"source":{"id":"2507.07313","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-09T22:22:49Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9ab6d82c54367a6f3f9080bcaed81477d3ff07dd06e710ff0ff514e798992138","abstract_canon_sha256":"d4caac9086db6ad8d66e0b03e54246fe861ed9f41b630a41b729f8a86cfadd56"},"schema_version":"1.0"},"canonical_sha256":"c2ef869199a00ae0e2f1ff4a246eeb526f451c1ecb08c66225062df19ce94f75","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:34:55.683576Z","signature_b64":"AW+YWOLBokM9DA50QZ+P9ezF9gFDMBAdTc8yWMaHM5n30UQ87ReqAaBIrSpyQDK+I/dPv6Bs3cRhug1yYZhfCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2ef869199a00ae0e2f1ff4a246eeb526f451c1ecb08c66225062df19ce94f75","last_reissued_at":"2026-07-05T11:34:55.683099Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:34:55.683099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.07313","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:34:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CRkRJPRcws1Dpltk84YGKzp0zKLYIJfaFAcwnd5+GjaknxPF3UxBMCHY2UuUBV6/3g17M1czyo8XDn/5IzIZDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T19:55:52.015607Z"},"content_sha256":"ebc0e7f1a97facccac8c73756ef3f00134fc1c763539770a53abb688980c1e3c","schema_version":"1.0","event_id":"sha256:ebc0e7f1a97facccac8c73756ef3f00134fc1c763539770a53abb688980c1e3c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YLXYNEMZUAFOBYXR75FCI3XLKJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Frontier LLMs Still Struggle with Simple Reasoning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alan Malek, Andr\\'as Gy\\\"orgy, Chi Jin, Csaba Szepesv\\'ari, Jiawei Ge, Nevena Lazic","submitted_at":"2025-07-09T22:22:49Z","abstract_excerpt":"While state-of-the-art large language models (LLMs) demonstrate advanced reasoning capabilities-achieving remarkable performance on challenging competitive math and coding benchmarks-they also frequently fail on tasks that are easy for humans. This work studies the performance of frontier LLMs on a broad set of such \"easy\" reasoning problems. By extending previous work in the literature, we create a suite of procedurally generated simple reasoning tasks, including counting, first-order logic, proof trees, and travel planning, with changeable parameters (such as document length. or the number o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07313","kind":"arxiv","version":1},"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/2507.07313/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":"2026-07-05T11:34:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dZCUalw0iyMKhScyS8j2IsFba1Vf3EW/uaLovHq1bPsoG56YGrvAvJwZjB0WH6+RDmpTpP9LqrWwIBGWxikhDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T19:55:52.016118Z"},"content_sha256":"e3bb9827f60bbf090a6658f46b7a5091dd0c0b211a86a809e206e17ed7723337","schema_version":"1.0","event_id":"sha256:e3bb9827f60bbf090a6658f46b7a5091dd0c0b211a86a809e206e17ed7723337"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ/bundle.json","state_url":"https://pith.science/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ/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":"2026-08-05T19:55:52Z","links":{"resolver":"https://pith.science/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ","bundle":"https://pith.science/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ/bundle.json","state":"https://pith.science/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YLXYNEMZUAFOBYXR75FCI3XLKJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YLXYNEMZUAFOBYXR75FCI3XLKJ","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":"d4caac9086db6ad8d66e0b03e54246fe861ed9f41b630a41b729f8a86cfadd56","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-09T22:22:49Z","title_canon_sha256":"9ab6d82c54367a6f3f9080bcaed81477d3ff07dd06e710ff0ff514e798992138"},"schema_version":"1.0","source":{"id":"2507.07313","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07313","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07313v1","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07313","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_12","alias_value":"YLXYNEMZUAFO","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_16","alias_value":"YLXYNEMZUAFOBYXR","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_8","alias_value":"YLXYNEMZ","created_at":"2026-07-05T11:34:55Z"}],"graph_snapshots":[{"event_id":"sha256:e3bb9827f60bbf090a6658f46b7a5091dd0c0b211a86a809e206e17ed7723337","target":"graph","created_at":"2026-07-05T11:34:55Z","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/2507.07313/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While state-of-the-art large language models (LLMs) demonstrate advanced reasoning capabilities-achieving remarkable performance on challenging competitive math and coding benchmarks-they also frequently fail on tasks that are easy for humans. This work studies the performance of frontier LLMs on a broad set of such \"easy\" reasoning problems. By extending previous work in the literature, we create a suite of procedurally generated simple reasoning tasks, including counting, first-order logic, proof trees, and travel planning, with changeable parameters (such as document length. or the number o","authors_text":"Alan Malek, Andr\\'as Gy\\\"orgy, Chi Jin, Csaba Szepesv\\'ari, Jiawei Ge, Nevena Lazic","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-09T22:22:49Z","title":"Frontier LLMs Still Struggle with Simple Reasoning Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07313","kind":"arxiv","version":1},"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:ebc0e7f1a97facccac8c73756ef3f00134fc1c763539770a53abb688980c1e3c","target":"record","created_at":"2026-07-05T11:34:55Z","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":"d4caac9086db6ad8d66e0b03e54246fe861ed9f41b630a41b729f8a86cfadd56","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-09T22:22:49Z","title_canon_sha256":"9ab6d82c54367a6f3f9080bcaed81477d3ff07dd06e710ff0ff514e798992138"},"schema_version":"1.0","source":{"id":"2507.07313","kind":"arxiv","version":1}},"canonical_sha256":"c2ef869199a00ae0e2f1ff4a246eeb526f451c1ecb08c66225062df19ce94f75","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c2ef869199a00ae0e2f1ff4a246eeb526f451c1ecb08c66225062df19ce94f75","first_computed_at":"2026-07-05T11:34:55.683099Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:34:55.683099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AW+YWOLBokM9DA50QZ+P9ezF9gFDMBAdTc8yWMaHM5n30UQ87ReqAaBIrSpyQDK+I/dPv6Bs3cRhug1yYZhfCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:34:55.683576Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.07313","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ebc0e7f1a97facccac8c73756ef3f00134fc1c763539770a53abb688980c1e3c","sha256:e3bb9827f60bbf090a6658f46b7a5091dd0c0b211a86a809e206e17ed7723337"],"state_sha256":"b81eabf66a2f9ff581799638b96951b413c29d6206d407037b565ddc7ddfe980"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vvG0P+/ZPkwp/XJVQtEvkCXnmdOLFaxvUDn8HAAtFlVXqNK4ixpHBj9QSaQabOKuuphUGe6EA+crVABlfoQMBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T19:55:52.021288Z","bundle_sha256":"2a61c2430e13913db8b23286324a057eefcaba7396ebd644e12def797e7df05f"}}