{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:H2Q4RL7CGGLWSYRDDEPGFA45DB","short_pith_number":"pith:H2Q4RL7C","schema_version":"1.0","canonical_sha256":"3ea1c8afe23197696223191e62839d185f4f1f7e491307129d6ca1e533159f44","source":{"kind":"arxiv","id":"2507.01551","version":2},"attestation_state":"computed","paper":{"title":"Self-Guided Process Reward Optimization with Redefined Step-wise Advantage for Process Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Hao Kong, Jing Tang, Lei Chen, Shuxian Liang, Wu Fei, Xiansheng Hua, Yang Lin, Yibo Yang","submitted_at":"2025-07-02T10:05:14Z","abstract_excerpt":"Process Reinforcement Learning~(PRL) has demonstrated considerable potential in enhancing the reasoning capabilities of Large Language Models~(LLMs). However, introducing additional process reward models incurs substantial computational overhead, and there is no unified theoretical framework for process-level advantage estimation. To bridge this gap, we propose \\textbf{S}elf-Guided \\textbf{P}rocess \\textbf{R}eward \\textbf{O}ptimization~(\\textbf{SPRO}), a novel framework that enables process-aware RL through two key innovations: (1) we first theoretically demonstrate that process rewards can be"},"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":"2507.01551","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-02T10:05:14Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"d910e7451f7a0ad662613f6561b706b8e0670e14b886a6a6ac1d493b0afd8f66","abstract_canon_sha256":"a83905eafabde9356dfcdcef2e98940ce998b9886d01ea7ff98a13192af04855"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:29.377154Z","signature_b64":"m0s+43e8v6ViDeRa2szaVH83eW5C0qG8PZS6++Ahn7kZUIngl4+0j4cbYD/A64o/tQdmKel0pRfDJecUNrSEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ea1c8afe23197696223191e62839d185f4f1f7e491307129d6ca1e533159f44","last_reissued_at":"2026-07-05T11:31:29.376690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:29.376690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Guided Process Reward Optimization with Redefined Step-wise Advantage for Process Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Hao Kong, Jing Tang, Lei Chen, Shuxian Liang, Wu Fei, Xiansheng Hua, Yang Lin, Yibo Yang","submitted_at":"2025-07-02T10:05:14Z","abstract_excerpt":"Process Reinforcement Learning~(PRL) has demonstrated considerable potential in enhancing the reasoning capabilities of Large Language Models~(LLMs). However, introducing additional process reward models incurs substantial computational overhead, and there is no unified theoretical framework for process-level advantage estimation. To bridge this gap, we propose \\textbf{S}elf-Guided \\textbf{P}rocess \\textbf{R}eward \\textbf{O}ptimization~(\\textbf{SPRO}), a novel framework that enables process-aware RL through two key innovations: (1) we first theoretically demonstrate that process rewards can be"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01551","kind":"arxiv","version":2},"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/2507.01551/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":"2507.01551","created_at":"2026-07-05T11:31:29.376750+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.01551v2","created_at":"2026-07-05T11:31:29.376750+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01551","created_at":"2026-07-05T11:31:29.376750+00:00"},{"alias_kind":"pith_short_12","alias_value":"H2Q4RL7CGGLW","created_at":"2026-07-05T11:31:29.376750+00:00"},{"alias_kind":"pith_short_16","alias_value":"H2Q4RL7CGGLWSYRD","created_at":"2026-07-05T11:31:29.376750+00:00"},{"alias_kind":"pith_short_8","alias_value":"H2Q4RL7C","created_at":"2026-07-05T11:31:29.376750+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":8,"sample":[{"citing_arxiv_id":"2606.09871","citing_title":"SD-GRPO: Verifiable Segment Decomposition for Long-Form Vision-Language Generation","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01249","citing_title":"Trust Region On-Policy Distillation","ref_index":131,"is_internal_anchor":true},{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":138,"is_internal_anchor":true},{"citing_arxiv_id":"2512.04949","citing_title":"CARL: Criticality-Aware Agentic Reinforcement Learning","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2601.04731","citing_title":"Miner:Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2503.09567","citing_title":"Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models","ref_index":179,"is_internal_anchor":true},{"citing_arxiv_id":"2604.13197","citing_title":"Unleashing Implicit Rewards: Prefix-Value Learning for Distribution-Level Optimization","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09459","citing_title":"From Reasoning to Agentic: Credit Assignment in Reinforcement Learning for Large Language Models","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB","json":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB.json","graph_json":"https://pith.science/api/pith-number/H2Q4RL7CGGLWSYRDDEPGFA45DB/graph.json","events_json":"https://pith.science/api/pith-number/H2Q4RL7CGGLWSYRDDEPGFA45DB/events.json","paper":"https://pith.science/paper/H2Q4RL7C"},"agent_actions":{"view_html":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB","download_json":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB.json","view_paper":"https://pith.science/paper/H2Q4RL7C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.01551&json=true","fetch_graph":"https://pith.science/api/pith-number/H2Q4RL7CGGLWSYRDDEPGFA45DB/graph.json","fetch_events":"https://pith.science/api/pith-number/H2Q4RL7CGGLWSYRDDEPGFA45DB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB/action/storage_attestation","attest_author":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB/action/author_attestation","sign_citation":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB/action/citation_signature","submit_replication":"https://pith.science/pith/H2Q4RL7CGGLWSYRDDEPGFA45DB/action/replication_record"}},"created_at":"2026-07-05T11:31:29.376750+00:00","updated_at":"2026-07-05T11:31:29.376750+00:00"}