{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PUI5B57G2LDGY4CFQJ5U7VUZCN","short_pith_number":"pith:PUI5B57G","canonical_record":{"source":{"id":"2505.23474","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T14:26:53Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"75311f69f9ce308a9f85ca495ecc937eaf1835724c000db914fa3b86071eb456","abstract_canon_sha256":"573c8a955917172066861be5544e18bee20626d1efa10d888eda0bc4f642db1f"},"schema_version":"1.0"},"canonical_sha256":"7d11d0f7e6d2c66c7045827b4fd6991351846977c83877aeb443afd35008b77d","source":{"kind":"arxiv","id":"2505.23474","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23474","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23474v1","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23474","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_12","alias_value":"PUI5B57G2LDG","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_16","alias_value":"PUI5B57G2LDGY4CF","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_8","alias_value":"PUI5B57G","created_at":"2026-07-05T11:12:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PUI5B57G2LDGY4CFQJ5U7VUZCN","target":"record","payload":{"canonical_record":{"source":{"id":"2505.23474","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T14:26:53Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"75311f69f9ce308a9f85ca495ecc937eaf1835724c000db914fa3b86071eb456","abstract_canon_sha256":"573c8a955917172066861be5544e18bee20626d1efa10d888eda0bc4f642db1f"},"schema_version":"1.0"},"canonical_sha256":"7d11d0f7e6d2c66c7045827b4fd6991351846977c83877aeb443afd35008b77d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:02.392758Z","signature_b64":"bfjr79IIOLi5zIAZKbA1B6GXRC/eHdwLSexpIM7s5nBGlGWPqayUmtKLQNADP+qA+3wbGTyLQ6U8IcW8kDvhDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d11d0f7e6d2c66c7045827b4fd6991351846977c83877aeb443afd35008b77d","last_reissued_at":"2026-07-05T11:12:02.392168Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:02.392168Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.23474","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":"2026-07-05T11:12:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tgw6CD77wgpwiLgrUUSwSRN1dT6Oc3knG6cb9XV+dNEeQHxlFcilVqXzIQsMFam+Ynktr/cOgnQyqWF7dzllDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T14:17:10.385532Z"},"content_sha256":"1b456e075c4187d91c718d2a67f0631ed7f12f027e5be6648c40c53806456e1f","schema_version":"1.0","event_id":"sha256:1b456e075c4187d91c718d2a67f0631ed7f12f027e5be6648c40c53806456e1f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PUI5B57G2LDGY4CFQJ5U7VUZCN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Socratic-PRMBench: Benchmarking Process Reward Models with Systematic Reasoning Patterns","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Fei Huang, Haiyang Yu, Jun Zhao, Kang Liu, Shizhu He, Xiang Li, Xinghua Zhang, Yongbin Li, Ziyang Huang","submitted_at":"2025-05-29T14:26:53Z","abstract_excerpt":"Process Reward Models (PRMs) are crucial in complex reasoning and problem-solving tasks (e.g., LLM agents with long-horizon decision-making) by verifying the correctness of each intermediate reasoning step. In real-world scenarios, LLMs may apply various reasoning patterns (e.g., decomposition) to solve a problem, potentially suffering from errors under various reasoning patterns. Therefore, PRMs are required to identify errors under various reasoning patterns during the reasoning process. However, existing benchmarks mainly focus on evaluating PRMs with stepwise correctness, ignoring a system"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23474","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/2505.23474/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":"2026-07-05T11:12:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NWvlFoxim6T6I70CTOoDWHSumhYnUsH92cOO8WRZfxZ07X7oFGJevfJw+BM/OgeKSXSQMEyk2OH1nalSHsr1CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T14:17:10.386614Z"},"content_sha256":"6baa6205c85a4e35fc8851b52205603c2fdf8053fbc46726bc72b1f23cb401df","schema_version":"1.0","event_id":"sha256:6baa6205c85a4e35fc8851b52205603c2fdf8053fbc46726bc72b1f23cb401df"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN/bundle.json","state_url":"https://pith.science/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN/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":"2026-08-13T14:17:10Z","links":{"resolver":"https://pith.science/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN","bundle":"https://pith.science/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN/bundle.json","state":"https://pith.science/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PUI5B57G2LDGY4CFQJ5U7VUZCN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PUI5B57G2LDGY4CFQJ5U7VUZCN","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":"573c8a955917172066861be5544e18bee20626d1efa10d888eda0bc4f642db1f","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T14:26:53Z","title_canon_sha256":"75311f69f9ce308a9f85ca495ecc937eaf1835724c000db914fa3b86071eb456"},"schema_version":"1.0","source":{"id":"2505.23474","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23474","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23474v1","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23474","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_12","alias_value":"PUI5B57G2LDG","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_16","alias_value":"PUI5B57G2LDGY4CF","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_8","alias_value":"PUI5B57G","created_at":"2026-07-05T11:12:02Z"}],"graph_snapshots":[{"event_id":"sha256:6baa6205c85a4e35fc8851b52205603c2fdf8053fbc46726bc72b1f23cb401df","target":"graph","created_at":"2026-07-05T11:12:02Z","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/2505.23474/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Process Reward Models (PRMs) are crucial in complex reasoning and problem-solving tasks (e.g., LLM agents with long-horizon decision-making) by verifying the correctness of each intermediate reasoning step. In real-world scenarios, LLMs may apply various reasoning patterns (e.g., decomposition) to solve a problem, potentially suffering from errors under various reasoning patterns. Therefore, PRMs are required to identify errors under various reasoning patterns during the reasoning process. However, existing benchmarks mainly focus on evaluating PRMs with stepwise correctness, ignoring a system","authors_text":"Fei Huang, Haiyang Yu, Jun Zhao, Kang Liu, Shizhu He, Xiang Li, Xinghua Zhang, Yongbin Li, Ziyang Huang","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T14:26:53Z","title":"Socratic-PRMBench: Benchmarking Process Reward Models with Systematic Reasoning Patterns"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23474","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":"sha256:1b456e075c4187d91c718d2a67f0631ed7f12f027e5be6648c40c53806456e1f","target":"record","created_at":"2026-07-05T11:12:02Z","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":"573c8a955917172066861be5544e18bee20626d1efa10d888eda0bc4f642db1f","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T14:26:53Z","title_canon_sha256":"75311f69f9ce308a9f85ca495ecc937eaf1835724c000db914fa3b86071eb456"},"schema_version":"1.0","source":{"id":"2505.23474","kind":"arxiv","version":1}},"canonical_sha256":"7d11d0f7e6d2c66c7045827b4fd6991351846977c83877aeb443afd35008b77d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7d11d0f7e6d2c66c7045827b4fd6991351846977c83877aeb443afd35008b77d","first_computed_at":"2026-07-05T11:12:02.392168Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:02.392168Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bfjr79IIOLi5zIAZKbA1B6GXRC/eHdwLSexpIM7s5nBGlGWPqayUmtKLQNADP+qA+3wbGTyLQ6U8IcW8kDvhDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:02.392758Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.23474","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1b456e075c4187d91c718d2a67f0631ed7f12f027e5be6648c40c53806456e1f","sha256:6baa6205c85a4e35fc8851b52205603c2fdf8053fbc46726bc72b1f23cb401df"],"state_sha256":"d994735a8475c7065e7789548534c9eab1d6c40d02f0b1d6adb5aa0a0f23e3f0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sQg2WGCbm6OsxvX+b+tzv6TVgxruZtfwA0i6wXLn6EFUuiOcyBV4w+ESynU1/GCnAa4kCPglZUMCL6Ynljz0Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T14:17:10.406326Z","bundle_sha256":"2873b9da66b8dc7f3b05a1ee4680483e3ca298f14ea1cc4e11320c725be7ec38"}}