{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DYYRLCTBEN27F67AQUWLLIWY5G","short_pith_number":"pith:DYYRLCTB","canonical_record":{"source":{"id":"2412.06559","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-09T15:11:40Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"fefd676b10464cc12f97b8601f0539fd6ad9f82c92c5856b575102250bef41b7","abstract_canon_sha256":"45f46ca5b11b6c2c7854efd428af7d7a5c9d59d3fae93e3c017831769d2a4d55"},"schema_version":"1.0"},"canonical_sha256":"1e31158a612375f2fbe0852cb5a2d8e9b19a792bcc9035f4c7aa8653047708ff","source":{"kind":"arxiv","id":"2412.06559","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.06559","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"arxiv_version","alias_value":"2412.06559v4","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.06559","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"pith_short_12","alias_value":"DYYRLCTBEN27","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"pith_short_16","alias_value":"DYYRLCTBEN27F67A","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"pith_short_8","alias_value":"DYYRLCTB","created_at":"2026-07-05T11:09:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DYYRLCTBEN27F67AQUWLLIWY5G","target":"record","payload":{"canonical_record":{"source":{"id":"2412.06559","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-09T15:11:40Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"fefd676b10464cc12f97b8601f0539fd6ad9f82c92c5856b575102250bef41b7","abstract_canon_sha256":"45f46ca5b11b6c2c7854efd428af7d7a5c9d59d3fae93e3c017831769d2a4d55"},"schema_version":"1.0"},"canonical_sha256":"1e31158a612375f2fbe0852cb5a2d8e9b19a792bcc9035f4c7aa8653047708ff","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:44.355785Z","signature_b64":"sE+PQpso+Pl8FRw8yY4AVZraOE0IlAzChN/q1I0ZQrz6lqDc7CzYEwKh/IUwMLH4TY1+0ma4qNg4vDSBQodzAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e31158a612375f2fbe0852cb5a2d8e9b19a792bcc9035f4c7aa8653047708ff","last_reissued_at":"2026-07-05T11:09:44.355284Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:44.355284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.06559","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":"2026-07-05T11:09:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VzUSfu3oJL/IknddAWw2kozelJPALX8lDqBHzTpM9Wo1oIazPdExt2J09z3/pfBIMcJg+OwzXF5wZlaR+kBbAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:26:45.558281Z"},"content_sha256":"2eef0888a519501453d73ac65e171e2a9a34fa93ad0569715250253ef46b13b0","schema_version":"1.0","event_id":"sha256:2eef0888a519501453d73ac65e171e2a9a34fa93ad0569715250253ef46b13b0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DYYRLCTBEN27F67AQUWLLIWY5G","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ProcessBench: Identifying Process Errors in Mathematical Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Beichen Zhang, Bowen Yu, Chujie Zheng, Dayiheng Liu, Jingren Zhou, Junyang Lin, Keming Lu, Runji Lin, Zhenru Zhang","submitted_at":"2024-12-09T15:11:40Z","abstract_excerpt":"As language models regularly make mistakes when solving math problems, automated identification of errors in the reasoning process becomes increasingly significant for their scalable oversight. In this paper, we introduce ProcessBench for measuring the ability to identify erroneous steps in mathematical reasoning. It consists of 3,400 test cases, primarily focused on competition- and Olympiad-level math problems. Each test case contains a step-by-step solution with error location annotated by human experts. Models are required to identify the earliest step that contains an error, or conclude t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.06559","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/2412.06559/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":"2026-07-05T11:09:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NRJU97hCUZY56XbHJxzoaYn6fhpNl6aExTskGuSF9igW9ZejWTPCtzZ9Q+WlgMS6rgLq8cxqd9hoHbdRGDpOCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:26:45.558782Z"},"content_sha256":"ab64f80e288a5dadc0fb6a492c39f558fec0dfab558bb1b3dfa23f2dae1a7efc","schema_version":"1.0","event_id":"sha256:ab64f80e288a5dadc0fb6a492c39f558fec0dfab558bb1b3dfa23f2dae1a7efc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DYYRLCTBEN27F67AQUWLLIWY5G/bundle.json","state_url":"https://pith.science/pith/DYYRLCTBEN27F67AQUWLLIWY5G/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DYYRLCTBEN27F67AQUWLLIWY5G/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":"2026-08-04T21:26:45Z","links":{"resolver":"https://pith.science/pith/DYYRLCTBEN27F67AQUWLLIWY5G","bundle":"https://pith.science/pith/DYYRLCTBEN27F67AQUWLLIWY5G/bundle.json","state":"https://pith.science/pith/DYYRLCTBEN27F67AQUWLLIWY5G/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DYYRLCTBEN27F67AQUWLLIWY5G/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DYYRLCTBEN27F67AQUWLLIWY5G","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":"45f46ca5b11b6c2c7854efd428af7d7a5c9d59d3fae93e3c017831769d2a4d55","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-09T15:11:40Z","title_canon_sha256":"fefd676b10464cc12f97b8601f0539fd6ad9f82c92c5856b575102250bef41b7"},"schema_version":"1.0","source":{"id":"2412.06559","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.06559","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"arxiv_version","alias_value":"2412.06559v4","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.06559","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"pith_short_12","alias_value":"DYYRLCTBEN27","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"pith_short_16","alias_value":"DYYRLCTBEN27F67A","created_at":"2026-07-05T11:09:44Z"},{"alias_kind":"pith_short_8","alias_value":"DYYRLCTB","created_at":"2026-07-05T11:09:44Z"}],"graph_snapshots":[{"event_id":"sha256:ab64f80e288a5dadc0fb6a492c39f558fec0dfab558bb1b3dfa23f2dae1a7efc","target":"graph","created_at":"2026-07-05T11:09:44Z","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/2412.06559/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As language models regularly make mistakes when solving math problems, automated identification of errors in the reasoning process becomes increasingly significant for their scalable oversight. In this paper, we introduce ProcessBench for measuring the ability to identify erroneous steps in mathematical reasoning. It consists of 3,400 test cases, primarily focused on competition- and Olympiad-level math problems. Each test case contains a step-by-step solution with error location annotated by human experts. Models are required to identify the earliest step that contains an error, or conclude t","authors_text":"Beichen Zhang, Bowen Yu, Chujie Zheng, Dayiheng Liu, Jingren Zhou, Junyang Lin, Keming Lu, Runji Lin, Zhenru Zhang","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-09T15:11:40Z","title":"ProcessBench: Identifying Process Errors in Mathematical Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.06559","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":"sha256:2eef0888a519501453d73ac65e171e2a9a34fa93ad0569715250253ef46b13b0","target":"record","created_at":"2026-07-05T11:09:44Z","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":"45f46ca5b11b6c2c7854efd428af7d7a5c9d59d3fae93e3c017831769d2a4d55","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-09T15:11:40Z","title_canon_sha256":"fefd676b10464cc12f97b8601f0539fd6ad9f82c92c5856b575102250bef41b7"},"schema_version":"1.0","source":{"id":"2412.06559","kind":"arxiv","version":4}},"canonical_sha256":"1e31158a612375f2fbe0852cb5a2d8e9b19a792bcc9035f4c7aa8653047708ff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1e31158a612375f2fbe0852cb5a2d8e9b19a792bcc9035f4c7aa8653047708ff","first_computed_at":"2026-07-05T11:09:44.355284Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:44.355284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sE+PQpso+Pl8FRw8yY4AVZraOE0IlAzChN/q1I0ZQrz6lqDc7CzYEwKh/IUwMLH4TY1+0ma4qNg4vDSBQodzAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:44.355785Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.06559","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2eef0888a519501453d73ac65e171e2a9a34fa93ad0569715250253ef46b13b0","sha256:ab64f80e288a5dadc0fb6a492c39f558fec0dfab558bb1b3dfa23f2dae1a7efc"],"state_sha256":"a31b245b90e68d096d4fd76e1b1f36b16b6284e45d1f7c19f3e6f954cafbdd4d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sGPAtaZE1Ur6wggCcJCPh6AA2Q2QrgzHwBvssYpYZDMnXJLinkL/mURLHNag8VdHXBuU0sF2/D5ISwxnP+IfDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T21:26:45.562859Z","bundle_sha256":"0a42aa3b1e0b075586f6a06e96f43b2b606aa6a8e731a3a1ccd9030f9b11e380"}}