{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KHTE3SOAI2WFVAXOJIXBXTKIGG","short_pith_number":"pith:KHTE3SOA","canonical_record":{"source":{"id":"2412.08819","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-11T23:31:06Z","cross_cats_sorted":[],"title_canon_sha256":"5490f5597aa14c185f98959a462ed2a0b204271c4884a0f1156377cbfab7e6fb","abstract_canon_sha256":"237106f2d8e0fd6d8c5fa0054f20b4510bf3ec900875f9ea4630bee23a5abf7d"},"schema_version":"1.0"},"canonical_sha256":"51e64dc9c046ac5a82ee4a2e1bcd4831b8ac5a771abe852835610a7f2b5c582a","source":{"kind":"arxiv","id":"2412.08819","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.08819","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"arxiv_version","alias_value":"2412.08819v1","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.08819","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"pith_short_12","alias_value":"KHTE3SOAI2WF","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"pith_short_16","alias_value":"KHTE3SOAI2WFVAXO","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"pith_short_8","alias_value":"KHTE3SOA","created_at":"2026-07-05T09:48:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KHTE3SOAI2WFVAXOJIXBXTKIGG","target":"record","payload":{"canonical_record":{"source":{"id":"2412.08819","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-11T23:31:06Z","cross_cats_sorted":[],"title_canon_sha256":"5490f5597aa14c185f98959a462ed2a0b204271c4884a0f1156377cbfab7e6fb","abstract_canon_sha256":"237106f2d8e0fd6d8c5fa0054f20b4510bf3ec900875f9ea4630bee23a5abf7d"},"schema_version":"1.0"},"canonical_sha256":"51e64dc9c046ac5a82ee4a2e1bcd4831b8ac5a771abe852835610a7f2b5c582a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:02.193007Z","signature_b64":"J2a2r5ulu8WG+fqQ/2J6VZ9+idIiStluQucko1vUl0oCqtL+N//Q035DJFpWNf7EpzPumvmKA4Xq1q1b6fetAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51e64dc9c046ac5a82ee4a2e1bcd4831b8ac5a771abe852835610a7f2b5c582a","last_reissued_at":"2026-07-05T09:48:02.192409Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:02.192409Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.08819","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":"2026-07-05T09:48:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A6XAfGXT7LJ0hI5yKkUH938dy03k4O3jqpzN6SThVhF+Hhngns6Fn75TRnUWb91LRfBBMZM7sFWWMFhi//bkAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T11:50:24.284322Z"},"content_sha256":"599e4aaf0abb646f146134f0125dda97b34d1595ddae9edbbffaa30140826875","schema_version":"1.0","event_id":"sha256:599e4aaf0abb646f146134f0125dda97b34d1595ddae9edbbffaa30140826875"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KHTE3SOAI2WFVAXOJIXBXTKIGG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HARP: A challenging human-annotated math reasoning benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aaditya K. Singh, Albert S. Yue, DJ Strouse, Lovish Madaan, Ted Moskovitz","submitted_at":"2024-12-11T23:31:06Z","abstract_excerpt":"Math reasoning is becoming an ever increasing area of focus as we scale large language models. However, even the previously-toughest evals like MATH are now close to saturated by frontier models (90.0% for o1-mini and 86.5% for Gemini 1.5 Pro). We introduce HARP, Human Annotated Reasoning Problems (for Math), consisting of 5,409 problems from the US national math competitions (A(J)HSME, AMC, AIME, USA(J)MO). Of these, 4,780 have answers that are automatically check-able (with libraries such as SymPy). These problems range six difficulty levels, with frontier models performing relatively poorly"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.08819","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/2412.08819/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":"2026-07-05T09:48:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"075oGoRN+3lX5GhUu9EznaQCV+Y7YpbMNNyZRw8Xf3K1/D3fIpQQP3vP+K6B6R5WNmbVXgs7F+8JfeD6t9vVCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T11:50:24.284874Z"},"content_sha256":"52ba846b3464c177cfd53c6f871907cfb39f32161994548644e7a5db824ab08a","schema_version":"1.0","event_id":"sha256:52ba846b3464c177cfd53c6f871907cfb39f32161994548644e7a5db824ab08a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG/bundle.json","state_url":"https://pith.science/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG/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":"2026-08-07T11:50:24Z","links":{"resolver":"https://pith.science/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG","bundle":"https://pith.science/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG/bundle.json","state":"https://pith.science/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KHTE3SOAI2WFVAXOJIXBXTKIGG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KHTE3SOAI2WFVAXOJIXBXTKIGG","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":"237106f2d8e0fd6d8c5fa0054f20b4510bf3ec900875f9ea4630bee23a5abf7d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-11T23:31:06Z","title_canon_sha256":"5490f5597aa14c185f98959a462ed2a0b204271c4884a0f1156377cbfab7e6fb"},"schema_version":"1.0","source":{"id":"2412.08819","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.08819","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"arxiv_version","alias_value":"2412.08819v1","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.08819","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"pith_short_12","alias_value":"KHTE3SOAI2WF","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"pith_short_16","alias_value":"KHTE3SOAI2WFVAXO","created_at":"2026-07-05T09:48:02Z"},{"alias_kind":"pith_short_8","alias_value":"KHTE3SOA","created_at":"2026-07-05T09:48:02Z"}],"graph_snapshots":[{"event_id":"sha256:52ba846b3464c177cfd53c6f871907cfb39f32161994548644e7a5db824ab08a","target":"graph","created_at":"2026-07-05T09:48:02Z","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/2412.08819/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Math reasoning is becoming an ever increasing area of focus as we scale large language models. However, even the previously-toughest evals like MATH are now close to saturated by frontier models (90.0% for o1-mini and 86.5% for Gemini 1.5 Pro). We introduce HARP, Human Annotated Reasoning Problems (for Math), consisting of 5,409 problems from the US national math competitions (A(J)HSME, AMC, AIME, USA(J)MO). Of these, 4,780 have answers that are automatically check-able (with libraries such as SymPy). These problems range six difficulty levels, with frontier models performing relatively poorly","authors_text":"Aaditya K. Singh, Albert S. Yue, DJ Strouse, Lovish Madaan, Ted Moskovitz","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-11T23:31:06Z","title":"HARP: A challenging human-annotated math reasoning benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.08819","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":"sha256:599e4aaf0abb646f146134f0125dda97b34d1595ddae9edbbffaa30140826875","target":"record","created_at":"2026-07-05T09:48:02Z","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":"237106f2d8e0fd6d8c5fa0054f20b4510bf3ec900875f9ea4630bee23a5abf7d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-11T23:31:06Z","title_canon_sha256":"5490f5597aa14c185f98959a462ed2a0b204271c4884a0f1156377cbfab7e6fb"},"schema_version":"1.0","source":{"id":"2412.08819","kind":"arxiv","version":1}},"canonical_sha256":"51e64dc9c046ac5a82ee4a2e1bcd4831b8ac5a771abe852835610a7f2b5c582a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"51e64dc9c046ac5a82ee4a2e1bcd4831b8ac5a771abe852835610a7f2b5c582a","first_computed_at":"2026-07-05T09:48:02.192409Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:48:02.192409Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"J2a2r5ulu8WG+fqQ/2J6VZ9+idIiStluQucko1vUl0oCqtL+N//Q035DJFpWNf7EpzPumvmKA4Xq1q1b6fetAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:48:02.193007Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.08819","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:599e4aaf0abb646f146134f0125dda97b34d1595ddae9edbbffaa30140826875","sha256:52ba846b3464c177cfd53c6f871907cfb39f32161994548644e7a5db824ab08a"],"state_sha256":"67a50f2f9dfc68ca66f75c2c215e77c64a870d6921c5553a33e4ecec3a7fa1d0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"14h0/JAo9H2a1mNpe1Qsx/iv5A6eCfDJxxmdDGul1JwUoCCH0qz7Ww5ufff81W2A0MT4DLMxcu5KHFaxIi/OCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T11:50:24.292144Z","bundle_sha256":"e990905ff4f4b41633dadcbc2cf008a837ad812cd4953fe5e4ab426c9921ee25"}}