{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:7MRDW5W4XKWX5HGCD43AGDBXWA","short_pith_number":"pith:7MRDW5W4","canonical_record":{"source":{"id":"2502.06453","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-10T13:31:46Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4bff9c2cf11e1c428219132ad0e8012386021ef1d332a86ebc99ce975b207f00","abstract_canon_sha256":"675a0a59efe37f45694118de3daaf8fbc4903c665cb2ffc3f03df8ff3a6c32fe"},"schema_version":"1.0"},"canonical_sha256":"fb223b76dcbaad7e9cc21f36030c37b0360c3e5894f4f802ac8f44ef9cc02e80","source":{"kind":"arxiv","id":"2502.06453","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06453","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06453v2","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06453","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"7MRDW5W4XKWX","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"7MRDW5W4XKWX5HGC","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"7MRDW5W4","created_at":"2026-07-05T10:13:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:7MRDW5W4XKWX5HGCD43AGDBXWA","target":"record","payload":{"canonical_record":{"source":{"id":"2502.06453","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-10T13:31:46Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4bff9c2cf11e1c428219132ad0e8012386021ef1d332a86ebc99ce975b207f00","abstract_canon_sha256":"675a0a59efe37f45694118de3daaf8fbc4903c665cb2ffc3f03df8ff3a6c32fe"},"schema_version":"1.0"},"canonical_sha256":"fb223b76dcbaad7e9cc21f36030c37b0360c3e5894f4f802ac8f44ef9cc02e80","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:27.799829Z","signature_b64":"muzN0P8QXdsjDraU2aoA7HcikkwHggEXZ5wwGHbQnlLV+1Ve6sr9lWUIzJjN7pMiA1uBHJJeB9pHzO6EfWRJBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb223b76dcbaad7e9cc21f36030c37b0360c3e5894f4f802ac8f44ef9cc02e80","last_reissued_at":"2026-07-05T10:13:27.799337Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:27.799337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.06453","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":"2026-07-05T10:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PXLMMwFWouQ5Dk+IHcSUhth4oPChke82/bqIv3HY1t/NkJQH1gvLR/nyGRWY0tsRzoJU0tMGvdndq0sCB1QHDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:08:30.147923Z"},"content_sha256":"f7068d36e9a138a5b8475788e6f7db35d148d5273ece2cefe3ec3d89de7f9fea","schema_version":"1.0","event_id":"sha256:f7068d36e9a138a5b8475788e6f7db35d148d5273ece2cefe3ec3d89de7f9fea"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:7MRDW5W4XKWX5HGCD43AGDBXWA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Chi Jin, Chiyuan Zhang, Hui Yuan, Jiacheng Guo, Jiawei Ge, Kaixuan Huang, Mengdi Wang, Ming Yin, Runzhe Wang, Shange Tang, Tianle Cai, Wenzhe Li, Xiang Ji, Xinyun Chen, Yangsibo Huang, Yingqing Guo, Yue Wu, Zihao Li","submitted_at":"2025-02-10T13:31:46Z","abstract_excerpt":"Large language models have demonstrated impressive performance on challenging mathematical reasoning tasks, which has triggered the discussion of whether the performance is achieved by true reasoning capability or memorization. To investigate this question, prior work has constructed mathematical benchmarks when questions undergo simple perturbations -- modifications that still preserve the underlying reasoning patterns of the solutions. However, no work has explored hard perturbations, which fundamentally change the nature of the problem so that the original solution steps do not apply. To br"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06453","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/2502.06453/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":"2026-07-05T10:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mIRKHNIS++bIdskyhy4VwnUJ6xdRn94aYrRDujXFCjt7CJSZVdpJ9bVIYlbBB8mubqs5LLwxYIrUM+QRX1CjAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:08:30.148998Z"},"content_sha256":"b7a53882554d0c64576ea65b1483c2544d31f71f26b0444d6371e547f61931e8","schema_version":"1.0","event_id":"sha256:b7a53882554d0c64576ea65b1483c2544d31f71f26b0444d6371e547f61931e8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7MRDW5W4XKWX5HGCD43AGDBXWA/bundle.json","state_url":"https://pith.science/pith/7MRDW5W4XKWX5HGCD43AGDBXWA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7MRDW5W4XKWX5HGCD43AGDBXWA/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":"2026-08-04T12:08:30Z","links":{"resolver":"https://pith.science/pith/7MRDW5W4XKWX5HGCD43AGDBXWA","bundle":"https://pith.science/pith/7MRDW5W4XKWX5HGCD43AGDBXWA/bundle.json","state":"https://pith.science/pith/7MRDW5W4XKWX5HGCD43AGDBXWA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7MRDW5W4XKWX5HGCD43AGDBXWA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7MRDW5W4XKWX5HGCD43AGDBXWA","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":"675a0a59efe37f45694118de3daaf8fbc4903c665cb2ffc3f03df8ff3a6c32fe","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-10T13:31:46Z","title_canon_sha256":"4bff9c2cf11e1c428219132ad0e8012386021ef1d332a86ebc99ce975b207f00"},"schema_version":"1.0","source":{"id":"2502.06453","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06453","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06453v2","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06453","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"7MRDW5W4XKWX","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"7MRDW5W4XKWX5HGC","created_at":"2026-07-05T10:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"7MRDW5W4","created_at":"2026-07-05T10:13:27Z"}],"graph_snapshots":[{"event_id":"sha256:b7a53882554d0c64576ea65b1483c2544d31f71f26b0444d6371e547f61931e8","target":"graph","created_at":"2026-07-05T10:13:27Z","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/2502.06453/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models have demonstrated impressive performance on challenging mathematical reasoning tasks, which has triggered the discussion of whether the performance is achieved by true reasoning capability or memorization. To investigate this question, prior work has constructed mathematical benchmarks when questions undergo simple perturbations -- modifications that still preserve the underlying reasoning patterns of the solutions. However, no work has explored hard perturbations, which fundamentally change the nature of the problem so that the original solution steps do not apply. To br","authors_text":"Chi Jin, Chiyuan Zhang, Hui Yuan, Jiacheng Guo, Jiawei Ge, Kaixuan Huang, Mengdi Wang, Ming Yin, Runzhe Wang, Shange Tang, Tianle Cai, Wenzhe Li, Xiang Ji, Xinyun Chen, Yangsibo Huang, Yingqing Guo, Yue Wu, Zihao Li","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-10T13:31:46Z","title":"MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06453","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:f7068d36e9a138a5b8475788e6f7db35d148d5273ece2cefe3ec3d89de7f9fea","target":"record","created_at":"2026-07-05T10:13:27Z","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":"675a0a59efe37f45694118de3daaf8fbc4903c665cb2ffc3f03df8ff3a6c32fe","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-10T13:31:46Z","title_canon_sha256":"4bff9c2cf11e1c428219132ad0e8012386021ef1d332a86ebc99ce975b207f00"},"schema_version":"1.0","source":{"id":"2502.06453","kind":"arxiv","version":2}},"canonical_sha256":"fb223b76dcbaad7e9cc21f36030c37b0360c3e5894f4f802ac8f44ef9cc02e80","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fb223b76dcbaad7e9cc21f36030c37b0360c3e5894f4f802ac8f44ef9cc02e80","first_computed_at":"2026-07-05T10:13:27.799337Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:13:27.799337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"muzN0P8QXdsjDraU2aoA7HcikkwHggEXZ5wwGHbQnlLV+1Ve6sr9lWUIzJjN7pMiA1uBHJJeB9pHzO6EfWRJBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:13:27.799829Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.06453","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f7068d36e9a138a5b8475788e6f7db35d148d5273ece2cefe3ec3d89de7f9fea","sha256:b7a53882554d0c64576ea65b1483c2544d31f71f26b0444d6371e547f61931e8"],"state_sha256":"c074661af96c44773c26cb9e6dd253ed32807e9542fe45ea78dbedc430d4f94d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eMh16IQlf1TSwfB9iQrDJkUwkdUfp2XRr2b1weoNzpEzSLKF4F44BHAflqQnDyGWZnaVwszCEROCQTHXESxDDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T12:08:30.158402Z","bundle_sha256":"da82f1fc3cbb16ab607c7294407f36d255e0827e296a38fb9a579029bb3e9035"}}