{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:PWNZGQI46DAA3JDGJGDFBJWFEV","short_pith_number":"pith:PWNZGQI4","canonical_record":{"source":{"id":"2405.00332","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T05:52:05Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"99aa56fec77c1c0e6ca55815bf7eae7d2269ef7d222ec0b07fb9ced9d31ac0d4","abstract_canon_sha256":"bf05f2b667fd90f7735da877121a5ea617547691104587b6bc99d992d900c258"},"schema_version":"1.0"},"canonical_sha256":"7d9b93411cf0c00da466498650a6c5255688f79c3ccf356a7493801b3582eec8","source":{"kind":"arxiv","id":"2405.00332","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.00332","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"arxiv_version","alias_value":"2405.00332v4","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.00332","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"pith_short_12","alias_value":"PWNZGQI46DAA","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"pith_short_16","alias_value":"PWNZGQI46DAA3JDG","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"pith_short_8","alias_value":"PWNZGQI4","created_at":"2026-07-05T09:39:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:PWNZGQI46DAA3JDGJGDFBJWFEV","target":"record","payload":{"canonical_record":{"source":{"id":"2405.00332","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T05:52:05Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"99aa56fec77c1c0e6ca55815bf7eae7d2269ef7d222ec0b07fb9ced9d31ac0d4","abstract_canon_sha256":"bf05f2b667fd90f7735da877121a5ea617547691104587b6bc99d992d900c258"},"schema_version":"1.0"},"canonical_sha256":"7d9b93411cf0c00da466498650a6c5255688f79c3ccf356a7493801b3582eec8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:39:22.852673Z","signature_b64":"hn4j2bXVVuF3WpHORgXdKmBfdM80BsBFtuoVBG/QTu2IcH15Ueq+yoNnhy469fc9pYBN5rgN4TETrf73n80XCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d9b93411cf0c00da466498650a6c5255688f79c3ccf356a7493801b3582eec8","last_reissued_at":"2026-07-05T09:39:22.852176Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:39:22.852176Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.00332","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:39:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m8/KFoYCwTAVnTQuYSOdaspTHlK3JipeclfH7ZeObd7EVb2ZBLKUHwIzx2Ucfzxd0a4AfVq189Mjj4zzp0h5DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:59:21.267666Z"},"content_sha256":"01746957b6470fcf6d15e680db73d3d6aa483f426e581e43178de7e39b601049","schema_version":"1.0","event_id":"sha256:01746957b6470fcf6d15e680db73d3d6aa483f426e581e43178de7e39b601049"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:PWNZGQI46DAA3JDGJGDFBJWFEV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Careful Examination of Large Language Model Performance on Grade School Arithmetic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Catherine Wu, Charlotte Zhuang, Dean Lee, Dylan Slack, Hugh Zhang, Jeff Da, Michele Lunati, Pranav Raja, Qin Lyu, Russell Kaplan, Sean Hendryx, Summer Yue, Tiffany Zhao, Vaughn Robinson, Will Song","submitted_at":"2024-05-01T05:52:05Z","abstract_excerpt":"Large language models (LLMs) have achieved impressive success on many benchmarks for mathematical reasoning. However, there is growing concern that some of this performance actually reflects dataset contamination, where data closely resembling benchmark questions leaks into the training data, instead of true reasoning ability. To investigate this claim rigorously, we commission Grade School Math 1000 (GSM1k). GSM1k is designed to mirror the style and complexity of the established GSM8k benchmark, the gold standard for measuring elementary mathematical reasoning. We ensure that the two benchmar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.00332","kind":"arxiv","version":4},"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/2405.00332/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":"2026-07-05T09:39:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"en9ZQu5wAP2u9qJhmSk7bBxF8jjt634pnEeHE2quCyjy+JKGrHQyQIXliQxP4OQ84szt4OR3Z8Q4uTCzw7dFCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:59:21.268194Z"},"content_sha256":"f18d488af92e022d3e48c40a1c43fc3edd11646e911597a9fc96854bbb59395e","schema_version":"1.0","event_id":"sha256:f18d488af92e022d3e48c40a1c43fc3edd11646e911597a9fc96854bbb59395e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PWNZGQI46DAA3JDGJGDFBJWFEV/bundle.json","state_url":"https://pith.science/pith/PWNZGQI46DAA3JDGJGDFBJWFEV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PWNZGQI46DAA3JDGJGDFBJWFEV/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":"2026-08-06T09:59:21Z","links":{"resolver":"https://pith.science/pith/PWNZGQI46DAA3JDGJGDFBJWFEV","bundle":"https://pith.science/pith/PWNZGQI46DAA3JDGJGDFBJWFEV/bundle.json","state":"https://pith.science/pith/PWNZGQI46DAA3JDGJGDFBJWFEV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PWNZGQI46DAA3JDGJGDFBJWFEV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PWNZGQI46DAA3JDGJGDFBJWFEV","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":"bf05f2b667fd90f7735da877121a5ea617547691104587b6bc99d992d900c258","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T05:52:05Z","title_canon_sha256":"99aa56fec77c1c0e6ca55815bf7eae7d2269ef7d222ec0b07fb9ced9d31ac0d4"},"schema_version":"1.0","source":{"id":"2405.00332","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.00332","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"arxiv_version","alias_value":"2405.00332v4","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.00332","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"pith_short_12","alias_value":"PWNZGQI46DAA","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"pith_short_16","alias_value":"PWNZGQI46DAA3JDG","created_at":"2026-07-05T09:39:22Z"},{"alias_kind":"pith_short_8","alias_value":"PWNZGQI4","created_at":"2026-07-05T09:39:22Z"}],"graph_snapshots":[{"event_id":"sha256:f18d488af92e022d3e48c40a1c43fc3edd11646e911597a9fc96854bbb59395e","target":"graph","created_at":"2026-07-05T09:39:22Z","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/2405.00332/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have achieved impressive success on many benchmarks for mathematical reasoning. However, there is growing concern that some of this performance actually reflects dataset contamination, where data closely resembling benchmark questions leaks into the training data, instead of true reasoning ability. To investigate this claim rigorously, we commission Grade School Math 1000 (GSM1k). GSM1k is designed to mirror the style and complexity of the established GSM8k benchmark, the gold standard for measuring elementary mathematical reasoning. We ensure that the two benchmar","authors_text":"Catherine Wu, Charlotte Zhuang, Dean Lee, Dylan Slack, Hugh Zhang, Jeff Da, Michele Lunati, Pranav Raja, Qin Lyu, Russell Kaplan, Sean Hendryx, Summer Yue, Tiffany Zhao, Vaughn Robinson, Will Song","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T05:52:05Z","title":"A Careful Examination of Large Language Model Performance on Grade School Arithmetic"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.00332","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":"sha256:01746957b6470fcf6d15e680db73d3d6aa483f426e581e43178de7e39b601049","target":"record","created_at":"2026-07-05T09:39:22Z","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":"bf05f2b667fd90f7735da877121a5ea617547691104587b6bc99d992d900c258","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T05:52:05Z","title_canon_sha256":"99aa56fec77c1c0e6ca55815bf7eae7d2269ef7d222ec0b07fb9ced9d31ac0d4"},"schema_version":"1.0","source":{"id":"2405.00332","kind":"arxiv","version":4}},"canonical_sha256":"7d9b93411cf0c00da466498650a6c5255688f79c3ccf356a7493801b3582eec8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7d9b93411cf0c00da466498650a6c5255688f79c3ccf356a7493801b3582eec8","first_computed_at":"2026-07-05T09:39:22.852176Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:39:22.852176Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hn4j2bXVVuF3WpHORgXdKmBfdM80BsBFtuoVBG/QTu2IcH15Ueq+yoNnhy469fc9pYBN5rgN4TETrf73n80XCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:39:22.852673Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.00332","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:01746957b6470fcf6d15e680db73d3d6aa483f426e581e43178de7e39b601049","sha256:f18d488af92e022d3e48c40a1c43fc3edd11646e911597a9fc96854bbb59395e"],"state_sha256":"f921c43fd06135b153f1b28e65130091afd0dfb808618d08b007b1262f5ccebc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vCs8RQDxM9qRolPJ4LF+r87lMkngMLCvIKd422sKs2ZweVdf1J8xRvC+oZcRSYLR1N7U6rfggruxAfxHY2cZAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T09:59:21.271798Z","bundle_sha256":"cfe5ecf7d07ea203eaea40f960764a3c78ac700b0b05e5230da33cfaa9188543"}}