{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AZEVUHCKKZTM4J53KAJAD5ROBZ","short_pith_number":"pith:AZEVUHCK","schema_version":"1.0","canonical_sha256":"06495a1c4a5666ce27bb501201f62e0e466ba178eeb9477c0be4dff4b9138dca","source":{"kind":"arxiv","id":"2507.17849","version":1},"attestation_state":"computed","paper":{"title":"Dynamic and Generalizable Process Reward Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Qinyuan Cheng, Qiushi Sun, Xipeng Qiu, Xuanjing Huang, Zhangyue Yin, Zhiyuan Zeng","submitted_at":"2025-07-23T18:17:22Z","abstract_excerpt":"Process Reward Models (PRMs) are crucial for guiding Large Language Models (LLMs) in complex scenarios by providing dense reward signals. However, existing PRMs primarily rely on heuristic approaches, which struggle with cross-domain generalization. While LLM-as-judge has been proposed to provide generalized rewards, current research has focused mainly on feedback results, overlooking the meaningful guidance embedded within the text. Additionally, static and coarse-grained evaluation criteria struggle to adapt to complex process supervision. To tackle these challenges, we propose Dynamic and G"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2507.17849","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-23T18:17:22Z","cross_cats_sorted":[],"title_canon_sha256":"7a9f4ed57a2a980a934f0e3a9e84774b069ba39149b51ba3daade4386a1723df","abstract_canon_sha256":"b73fa9d9c9cf8134725dca4175e41170d8fc1e1c26c640b91f5bc614a17cf0a5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:36.275510Z","signature_b64":"JFRAzCOuZKW4wMIdqjmFB3KcoK9o0avtWmRn8NpAbd+UtwYZB+ax+n5DlDFAlfzYYL/fD7cSGWGD4mYG9w66CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06495a1c4a5666ce27bb501201f62e0e466ba178eeb9477c0be4dff4b9138dca","last_reissued_at":"2026-07-05T11:42:36.275039Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:36.275039Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic and Generalizable Process Reward Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Qinyuan Cheng, Qiushi Sun, Xipeng Qiu, Xuanjing Huang, Zhangyue Yin, Zhiyuan Zeng","submitted_at":"2025-07-23T18:17:22Z","abstract_excerpt":"Process Reward Models (PRMs) are crucial for guiding Large Language Models (LLMs) in complex scenarios by providing dense reward signals. However, existing PRMs primarily rely on heuristic approaches, which struggle with cross-domain generalization. While LLM-as-judge has been proposed to provide generalized rewards, current research has focused mainly on feedback results, overlooking the meaningful guidance embedded within the text. Additionally, static and coarse-grained evaluation criteria struggle to adapt to complex process supervision. To tackle these challenges, we propose Dynamic and G"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17849","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/2507.17849/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2507.17849","created_at":"2026-07-05T11:42:36.275096+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.17849v1","created_at":"2026-07-05T11:42:36.275096+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17849","created_at":"2026-07-05T11:42:36.275096+00:00"},{"alias_kind":"pith_short_12","alias_value":"AZEVUHCKKZTM","created_at":"2026-07-05T11:42:36.275096+00:00"},{"alias_kind":"pith_short_16","alias_value":"AZEVUHCKKZTM4J53","created_at":"2026-07-05T11:42:36.275096+00:00"},{"alias_kind":"pith_short_8","alias_value":"AZEVUHCK","created_at":"2026-07-05T11:42:36.275096+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.15951","citing_title":"From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level Grounding","ref_index":88,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ","json":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ.json","graph_json":"https://pith.science/api/pith-number/AZEVUHCKKZTM4J53KAJAD5ROBZ/graph.json","events_json":"https://pith.science/api/pith-number/AZEVUHCKKZTM4J53KAJAD5ROBZ/events.json","paper":"https://pith.science/paper/AZEVUHCK"},"agent_actions":{"view_html":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ","download_json":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ.json","view_paper":"https://pith.science/paper/AZEVUHCK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.17849&json=true","fetch_graph":"https://pith.science/api/pith-number/AZEVUHCKKZTM4J53KAJAD5ROBZ/graph.json","fetch_events":"https://pith.science/api/pith-number/AZEVUHCKKZTM4J53KAJAD5ROBZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ/action/storage_attestation","attest_author":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ/action/author_attestation","sign_citation":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ/action/citation_signature","submit_replication":"https://pith.science/pith/AZEVUHCKKZTM4J53KAJAD5ROBZ/action/replication_record"}},"created_at":"2026-07-05T11:42:36.275096+00:00","updated_at":"2026-07-05T11:42:36.275096+00:00"}