{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:VQ2A3KPYQHAX55NZOIGZBV3CX2","short_pith_number":"pith:VQ2A3KPY","canonical_record":{"source":{"id":"2402.14660","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T16:06:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b25d692a89e6433ed52c4c4ea82460209275b4a8b60fd100f679edae48cf4b64","abstract_canon_sha256":"2c8599789631bde5ffb069a06463b7aa01f74e15a90c611a77527ad7af653909"},"schema_version":"1.0"},"canonical_sha256":"ac340da9f881c17ef5b9720d90d762be8b55b053ba087ded9d903722bcd05d45","source":{"kind":"arxiv","id":"2402.14660","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14660","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14660v2","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14660","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"pith_short_12","alias_value":"VQ2A3KPYQHAX","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"pith_short_16","alias_value":"VQ2A3KPYQHAX55NZ","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"pith_short_8","alias_value":"VQ2A3KPY","created_at":"2026-07-05T07:48:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:VQ2A3KPYQHAX55NZOIGZBV3CX2","target":"record","payload":{"canonical_record":{"source":{"id":"2402.14660","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T16:06:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b25d692a89e6433ed52c4c4ea82460209275b4a8b60fd100f679edae48cf4b64","abstract_canon_sha256":"2c8599789631bde5ffb069a06463b7aa01f74e15a90c611a77527ad7af653909"},"schema_version":"1.0"},"canonical_sha256":"ac340da9f881c17ef5b9720d90d762be8b55b053ba087ded9d903722bcd05d45","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:35.422338Z","signature_b64":"pmBeop3dUV+4NVmovwGSXTKjnrlMBwF5O9Y6P4Ch2reCZH+ifp2zkqG++vqxo7xqwbr5FpGFrx600ByEhPIXDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac340da9f881c17ef5b9720d90d762be8b55b053ba087ded9d903722bcd05d45","last_reissued_at":"2026-07-05T07:48:35.421821Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:35.421821Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.14660","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":"2026-07-05T07:48:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iraklYLEc1B3R4/0HuXFtn45iSHKzLaMk2Wl4cUVfIuAmJfHwa5HcPBK2ZWRwu4vxQg5kqT5ls3Ijj61hi+5Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T14:47:16.892712Z"},"content_sha256":"20af455819326df3bfc0ffaaad821bff01797675977bbaf4d2b2ff22460a8410","schema_version":"1.0","event_id":"sha256:20af455819326df3bfc0ffaaad821bff01797675977bbaf4d2b2ff22460a8410"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:VQ2A3KPYQHAX55NZOIGZBV3CX2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ConceptMath: A Bilingual Concept-wise Benchmark for Measuring Mathematical Reasoning of Large Language Models","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Zheng, Chenchen Zhang, Haibin Chen, Jiaheng Liu, Jie Liu, Tiezheng Ge, Wanli Ouyang, Wenbo Su, Xingyuan Bu, Yanan Wu, Yuanxing Zhang, Zhanhui Zhou, Zhiqi Bai","submitted_at":"2024-02-22T16:06:49Z","abstract_excerpt":"This paper introduces ConceptMath, a bilingual (English and Chinese), fine-grained benchmark that evaluates concept-wise mathematical reasoning of Large Language Models (LLMs). Unlike traditional benchmarks that evaluate general mathematical reasoning with an average accuracy, ConceptMath systematically organizes math problems under a hierarchy of math concepts, so that mathematical reasoning can be evaluated at different granularity with concept-wise accuracies. Based on our ConcepthMath, we evaluate a broad range of LLMs, and we observe existing LLMs, though achieving high average accuracies"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14660","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/2402.14660/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":"2026-07-05T07:48:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"74VpIR6ykP3wE0nQWGgRawlV4fnDjb5NtMb4CRb5amhSf/tjHSQI556c9HfPESF/Ofc/ynuNw0wM46cfSEDSCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T14:47:16.893682Z"},"content_sha256":"e26438051d4ed78cde9d1e7d4e51d1f92481a9e76a017945efec65f33eff07fa","schema_version":"1.0","event_id":"sha256:e26438051d4ed78cde9d1e7d4e51d1f92481a9e76a017945efec65f33eff07fa"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2/bundle.json","state_url":"https://pith.science/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2/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":"2026-08-10T14:47:16Z","links":{"resolver":"https://pith.science/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2","bundle":"https://pith.science/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2/bundle.json","state":"https://pith.science/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VQ2A3KPYQHAX55NZOIGZBV3CX2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:VQ2A3KPYQHAX55NZOIGZBV3CX2","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":"2c8599789631bde5ffb069a06463b7aa01f74e15a90c611a77527ad7af653909","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T16:06:49Z","title_canon_sha256":"b25d692a89e6433ed52c4c4ea82460209275b4a8b60fd100f679edae48cf4b64"},"schema_version":"1.0","source":{"id":"2402.14660","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14660","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14660v2","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14660","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"pith_short_12","alias_value":"VQ2A3KPYQHAX","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"pith_short_16","alias_value":"VQ2A3KPYQHAX55NZ","created_at":"2026-07-05T07:48:35Z"},{"alias_kind":"pith_short_8","alias_value":"VQ2A3KPY","created_at":"2026-07-05T07:48:35Z"}],"graph_snapshots":[{"event_id":"sha256:e26438051d4ed78cde9d1e7d4e51d1f92481a9e76a017945efec65f33eff07fa","target":"graph","created_at":"2026-07-05T07:48:35Z","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/2402.14660/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces ConceptMath, a bilingual (English and Chinese), fine-grained benchmark that evaluates concept-wise mathematical reasoning of Large Language Models (LLMs). Unlike traditional benchmarks that evaluate general mathematical reasoning with an average accuracy, ConceptMath systematically organizes math problems under a hierarchy of math concepts, so that mathematical reasoning can be evaluated at different granularity with concept-wise accuracies. Based on our ConcepthMath, we evaluate a broad range of LLMs, and we observe existing LLMs, though achieving high average accuracies","authors_text":"Bo Zheng, Chenchen Zhang, Haibin Chen, Jiaheng Liu, Jie Liu, Tiezheng Ge, Wanli Ouyang, Wenbo Su, Xingyuan Bu, Yanan Wu, Yuanxing Zhang, Zhanhui Zhou, Zhiqi Bai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T16:06:49Z","title":"ConceptMath: A Bilingual Concept-wise Benchmark for Measuring Mathematical Reasoning of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14660","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":"sha256:20af455819326df3bfc0ffaaad821bff01797675977bbaf4d2b2ff22460a8410","target":"record","created_at":"2026-07-05T07:48:35Z","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":"2c8599789631bde5ffb069a06463b7aa01f74e15a90c611a77527ad7af653909","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T16:06:49Z","title_canon_sha256":"b25d692a89e6433ed52c4c4ea82460209275b4a8b60fd100f679edae48cf4b64"},"schema_version":"1.0","source":{"id":"2402.14660","kind":"arxiv","version":2}},"canonical_sha256":"ac340da9f881c17ef5b9720d90d762be8b55b053ba087ded9d903722bcd05d45","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ac340da9f881c17ef5b9720d90d762be8b55b053ba087ded9d903722bcd05d45","first_computed_at":"2026-07-05T07:48:35.421821Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:48:35.421821Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pmBeop3dUV+4NVmovwGSXTKjnrlMBwF5O9Y6P4Ch2reCZH+ifp2zkqG++vqxo7xqwbr5FpGFrx600ByEhPIXDA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:48:35.422338Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.14660","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:20af455819326df3bfc0ffaaad821bff01797675977bbaf4d2b2ff22460a8410","sha256:e26438051d4ed78cde9d1e7d4e51d1f92481a9e76a017945efec65f33eff07fa"],"state_sha256":"dd3b08f340bbfd4aa88ada3af0b36c402529225db7793ca3e81ae9ededd1dcbb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"y4P1SmX75G/HuoML5XO3+QBCc581IR06Br4Z4UKzE89d5EpfYUUZwU1iNgaqQYxVTWq74ag4XvpbEC2cd+KSBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T14:47:16.900029Z","bundle_sha256":"8fa063cd5e4c37faf13c276ac3d91e331722da9f99a6818183a83cda87019340"}}