{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:DLM7TVE7P3X6Z6JM7BEHV6LZIN","short_pith_number":"pith:DLM7TVE7","canonical_record":{"source":{"id":"2503.21295","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-27T09:23:08Z","cross_cats_sorted":[],"title_canon_sha256":"820083602dbad970920a2ad2bd86a7fd9d1359a2e325df242680fa7b9e36d6d5","abstract_canon_sha256":"7b89c429c7925abb1044b41fca21a91c865d6dad7a66f87346b13e6698b1a59e"},"schema_version":"1.0"},"canonical_sha256":"1ad9f9d49f7eefecf92cf8487af979437be51b33fd39e77f1fa2a467a6d053e4","source":{"kind":"arxiv","id":"2503.21295","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21295","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21295v1","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21295","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"pith_short_12","alias_value":"DLM7TVE7P3X6","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"pith_short_16","alias_value":"DLM7TVE7P3X6Z6JM","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"pith_short_8","alias_value":"DLM7TVE7","created_at":"2026-07-05T10:40:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:DLM7TVE7P3X6Z6JM7BEHV6LZIN","target":"record","payload":{"canonical_record":{"source":{"id":"2503.21295","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-27T09:23:08Z","cross_cats_sorted":[],"title_canon_sha256":"820083602dbad970920a2ad2bd86a7fd9d1359a2e325df242680fa7b9e36d6d5","abstract_canon_sha256":"7b89c429c7925abb1044b41fca21a91c865d6dad7a66f87346b13e6698b1a59e"},"schema_version":"1.0"},"canonical_sha256":"1ad9f9d49f7eefecf92cf8487af979437be51b33fd39e77f1fa2a467a6d053e4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:40:22.669532Z","signature_b64":"cc5+GVSDRYdLBM9I+pMOVNUrO92kRp14bIPvu45tVNCj17ArMo23rcYSeQAl8gXml3NZbST6ZreowdEKcbhnCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ad9f9d49f7eefecf92cf8487af979437be51b33fd39e77f1fa2a467a6d053e4","last_reissued_at":"2026-07-05T10:40:22.669035Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:40:22.669035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.21295","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":"2026-07-05T10:40:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d/g932sVcb3GHMzSfDYp2GUMK9PwiLbamIbpIxSTFASQNQzayVw/TtBgZyoqlCUyYVBXiwPsbNOY7z6PVWPiBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T14:28:06.607311Z"},"content_sha256":"1b2fdf6aad3385f6552de236336b4daa17871e3cbe2618d9f239efee21ecd321","schema_version":"1.0","event_id":"sha256:1b2fdf6aad3385f6552de236336b4daa17871e3cbe2618d9f239efee21ecd321"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:DLM7TVE7P3X6Z6JM7BEHV6LZIN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"R-PRM: Reasoning-Driven Process Reward Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiajun Chen, Junxiao Liu, Shuaijie She, Shujian Huang, Xin Huang, Yifeng Liu","submitted_at":"2025-03-27T09:23:08Z","abstract_excerpt":"Large language models (LLMs) inevitably make mistakes when performing step-by-step mathematical reasoning. Process Reward Models (PRMs) have emerged as a promising solution by evaluating each reasoning step. However, existing PRMs typically output evaluation scores directly, limiting both learning efficiency and evaluation accuracy, which is further exacerbated by the scarcity of annotated data. To address these issues, we propose Reasoning-Driven Process Reward Modeling (R-PRM). First, we leverage stronger LLMs to generate seed data from limited annotations, effectively bootstrapping our mode"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21295","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/2503.21295/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":"2026-07-05T10:40:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dmcPZqBwci2G/H96skbE77q6yf3kEJodrI3rnF0+A1utHjm/5rUKY6oKobgk1Nr7vqJbgzdL6yWBbLkcW1gaCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T14:28:06.607843Z"},"content_sha256":"2e8721b2804cb35a30ede617cacd035a502ad6ba4fa27a95e83b99d7d3a73127","schema_version":"1.0","event_id":"sha256:2e8721b2804cb35a30ede617cacd035a502ad6ba4fa27a95e83b99d7d3a73127"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN/bundle.json","state_url":"https://pith.science/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN/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":"2026-08-05T14:28:06Z","links":{"resolver":"https://pith.science/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN","bundle":"https://pith.science/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN/bundle.json","state":"https://pith.science/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DLM7TVE7P3X6Z6JM7BEHV6LZIN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DLM7TVE7P3X6Z6JM7BEHV6LZIN","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":"7b89c429c7925abb1044b41fca21a91c865d6dad7a66f87346b13e6698b1a59e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-27T09:23:08Z","title_canon_sha256":"820083602dbad970920a2ad2bd86a7fd9d1359a2e325df242680fa7b9e36d6d5"},"schema_version":"1.0","source":{"id":"2503.21295","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21295","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21295v1","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21295","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"pith_short_12","alias_value":"DLM7TVE7P3X6","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"pith_short_16","alias_value":"DLM7TVE7P3X6Z6JM","created_at":"2026-07-05T10:40:22Z"},{"alias_kind":"pith_short_8","alias_value":"DLM7TVE7","created_at":"2026-07-05T10:40:22Z"}],"graph_snapshots":[{"event_id":"sha256:2e8721b2804cb35a30ede617cacd035a502ad6ba4fa27a95e83b99d7d3a73127","target":"graph","created_at":"2026-07-05T10:40:22Z","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/2503.21295/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) inevitably make mistakes when performing step-by-step mathematical reasoning. Process Reward Models (PRMs) have emerged as a promising solution by evaluating each reasoning step. However, existing PRMs typically output evaluation scores directly, limiting both learning efficiency and evaluation accuracy, which is further exacerbated by the scarcity of annotated data. To address these issues, we propose Reasoning-Driven Process Reward Modeling (R-PRM). First, we leverage stronger LLMs to generate seed data from limited annotations, effectively bootstrapping our mode","authors_text":"Jiajun Chen, Junxiao Liu, Shuaijie She, Shujian Huang, Xin Huang, Yifeng Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-27T09:23:08Z","title":"R-PRM: Reasoning-Driven Process Reward Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21295","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":"sha256:1b2fdf6aad3385f6552de236336b4daa17871e3cbe2618d9f239efee21ecd321","target":"record","created_at":"2026-07-05T10:40:22Z","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":"7b89c429c7925abb1044b41fca21a91c865d6dad7a66f87346b13e6698b1a59e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-27T09:23:08Z","title_canon_sha256":"820083602dbad970920a2ad2bd86a7fd9d1359a2e325df242680fa7b9e36d6d5"},"schema_version":"1.0","source":{"id":"2503.21295","kind":"arxiv","version":1}},"canonical_sha256":"1ad9f9d49f7eefecf92cf8487af979437be51b33fd39e77f1fa2a467a6d053e4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1ad9f9d49f7eefecf92cf8487af979437be51b33fd39e77f1fa2a467a6d053e4","first_computed_at":"2026-07-05T10:40:22.669035Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:40:22.669035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cc5+GVSDRYdLBM9I+pMOVNUrO92kRp14bIPvu45tVNCj17ArMo23rcYSeQAl8gXml3NZbST6ZreowdEKcbhnCg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:40:22.669532Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.21295","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1b2fdf6aad3385f6552de236336b4daa17871e3cbe2618d9f239efee21ecd321","sha256:2e8721b2804cb35a30ede617cacd035a502ad6ba4fa27a95e83b99d7d3a73127"],"state_sha256":"143bb723bc29ef26b88afd3cacba9b8ac90b804b34ff4353e4d7e2bd512e2ff0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VdGW5v1eOmMsVm1WRCpGuGUPJN2TH+5nbmWqcqVOOdEh0txk0qqHktQtromasAkMUCEX5kkzHznatOZlg6FQAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T14:28:06.613095Z","bundle_sha256":"c1a0ff4c9a6cd15c55123ee89493ec2fb792435f54489df7d4350ad53f944343"}}