{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:GNOCPX6BGSRCEMIOGR4MNSS2RQ","short_pith_number":"pith:GNOCPX6B","canonical_record":{"source":{"id":"2602.07832","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-08T05:47:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"509a323776fed6f653fed2861c8f25c96cb7c38854b1ab726bc74d56f2a41cca","abstract_canon_sha256":"007badd1aec1adbe14c5e4652294374b3e5e37de24771f7a013f3834f4a4f670"},"schema_version":"1.0"},"canonical_sha256":"335c27dfc134a222310e3478c6ca5a8c1557472d21508f64ab701306964383b1","source":{"kind":"arxiv","id":"2602.07832","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.07832","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"arxiv_version","alias_value":"2602.07832v2","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.07832","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"pith_short_12","alias_value":"GNOCPX6BGSRC","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"pith_short_16","alias_value":"GNOCPX6BGSRCEMIO","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"pith_short_8","alias_value":"GNOCPX6B","created_at":"2026-05-21T01:04:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:GNOCPX6BGSRCEMIOGR4MNSS2RQ","target":"record","payload":{"canonical_record":{"source":{"id":"2602.07832","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-08T05:47:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"509a323776fed6f653fed2861c8f25c96cb7c38854b1ab726bc74d56f2a41cca","abstract_canon_sha256":"007badd1aec1adbe14c5e4652294374b3e5e37de24771f7a013f3834f4a4f670"},"schema_version":"1.0"},"canonical_sha256":"335c27dfc134a222310e3478c6ca5a8c1557472d21508f64ab701306964383b1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:22.550953Z","signature_b64":"jgqxdW8McupCaRResQxSR0/5fmxc1hcFvGbT1JHc4S9X/M+lo1x3bru2aU7PI1rdSZTPTmkQ9ruVahmO1+VRBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"335c27dfc134a222310e3478c6ca5a8c1557472d21508f64ab701306964383b1","last_reissued_at":"2026-05-21T01:04:22.550047Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:22.550047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.07832","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-21T01:04:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"R4tpj3KvH/TRo+45ZF4CX4nZ0JWGviVbGuFSzjgB2Cm0dmxGvCK3pCyZIS7DAfKxxdEV0cVayTpCyGLFDge4CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T07:57:52.109210Z"},"content_sha256":"d7164c9be2f67fee118f1b6386b9670c71f8931b8a6d70d2beb6daaeb6b6ea6e","schema_version":"1.0","event_id":"sha256:d7164c9be2f67fee118f1b6386b9670c71f8931b8a6d70d2beb6daaeb6b6ea6e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:GNOCPX6BGSRCEMIOGR4MNSS2RQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"rePIRL: Learn PRM with Inverse RL for LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Kaijie Zhu, Lun Wang, Wenbo Guo, Xian Wu, Ying Zhang","submitted_at":"2026-02-08T05:47:27Z","abstract_excerpt":"Process rewards have been widely used in deep reinforcement learning to improve training efficiency, reduce variance, and prevent reward hacking. In LLM reasoning, existing works also explore various solutions for learning effective process reward models (PRM) with or without the help of an expert policy. However, existing methods either rely on strong assumptions about the expert policies (e.g., requiring their reward functions) or suffer intrinsic limitations (e.g., entropy collapse), resulting in weak PRMs or limited generalizability. In this paper, we introduce rePIRL, an inverse RL-inspir"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.07832","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/2602.07832/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":"2026-05-21T01:04:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mjxMFejX66zWBBYG0z3ifFayQWp30+KjlsADqiR95LWfiXkj6hlSYjPPglgFojBF+wo2Hhh8DyBnekfEcBw5CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T07:57:52.110000Z"},"content_sha256":"216054945704498e4cbe4bbc62c452e8d1fdc80ce80e0b4dbd9510f2a4fdcfcd","schema_version":"1.0","event_id":"sha256:216054945704498e4cbe4bbc62c452e8d1fdc80ce80e0b4dbd9510f2a4fdcfcd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ/bundle.json","state_url":"https://pith.science/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ/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":"2026-05-25T07:57:52Z","links":{"resolver":"https://pith.science/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ","bundle":"https://pith.science/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ/bundle.json","state":"https://pith.science/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GNOCPX6BGSRCEMIOGR4MNSS2RQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GNOCPX6BGSRCEMIOGR4MNSS2RQ","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":"007badd1aec1adbe14c5e4652294374b3e5e37de24771f7a013f3834f4a4f670","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-08T05:47:27Z","title_canon_sha256":"509a323776fed6f653fed2861c8f25c96cb7c38854b1ab726bc74d56f2a41cca"},"schema_version":"1.0","source":{"id":"2602.07832","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.07832","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"arxiv_version","alias_value":"2602.07832v2","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.07832","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"pith_short_12","alias_value":"GNOCPX6BGSRC","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"pith_short_16","alias_value":"GNOCPX6BGSRCEMIO","created_at":"2026-05-21T01:04:22Z"},{"alias_kind":"pith_short_8","alias_value":"GNOCPX6B","created_at":"2026-05-21T01:04:22Z"}],"graph_snapshots":[{"event_id":"sha256:216054945704498e4cbe4bbc62c452e8d1fdc80ce80e0b4dbd9510f2a4fdcfcd","target":"graph","created_at":"2026-05-21T01:04:22Z","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/2602.07832/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Process rewards have been widely used in deep reinforcement learning to improve training efficiency, reduce variance, and prevent reward hacking. In LLM reasoning, existing works also explore various solutions for learning effective process reward models (PRM) with or without the help of an expert policy. However, existing methods either rely on strong assumptions about the expert policies (e.g., requiring their reward functions) or suffer intrinsic limitations (e.g., entropy collapse), resulting in weak PRMs or limited generalizability. In this paper, we introduce rePIRL, an inverse RL-inspir","authors_text":"Kaijie Zhu, Lun Wang, Wenbo Guo, Xian Wu, Ying Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-08T05:47:27Z","title":"rePIRL: Learn PRM with Inverse RL for LLM Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.07832","kind":"arxiv","version":2},"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:d7164c9be2f67fee118f1b6386b9670c71f8931b8a6d70d2beb6daaeb6b6ea6e","target":"record","created_at":"2026-05-21T01:04:22Z","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":"007badd1aec1adbe14c5e4652294374b3e5e37de24771f7a013f3834f4a4f670","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-08T05:47:27Z","title_canon_sha256":"509a323776fed6f653fed2861c8f25c96cb7c38854b1ab726bc74d56f2a41cca"},"schema_version":"1.0","source":{"id":"2602.07832","kind":"arxiv","version":2}},"canonical_sha256":"335c27dfc134a222310e3478c6ca5a8c1557472d21508f64ab701306964383b1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"335c27dfc134a222310e3478c6ca5a8c1557472d21508f64ab701306964383b1","first_computed_at":"2026-05-21T01:04:22.550047Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:22.550047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jgqxdW8McupCaRResQxSR0/5fmxc1hcFvGbT1JHc4S9X/M+lo1x3bru2aU7PI1rdSZTPTmkQ9ruVahmO1+VRBQ==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:22.550953Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.07832","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d7164c9be2f67fee118f1b6386b9670c71f8931b8a6d70d2beb6daaeb6b6ea6e","sha256:216054945704498e4cbe4bbc62c452e8d1fdc80ce80e0b4dbd9510f2a4fdcfcd"],"state_sha256":"7b411acd8614da2cbb4b561018ee7741df6d6a5d556bc9af264e2d5a4026159a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sIyOVocuxjjY1QUzFPSo0WLFCOKXGCRRNIDwtHrgGpVue0631uWMtJAA3q9/+2hel0ifr8pNYTEP5qBJ9Af0AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T07:57:52.113467Z","bundle_sha256":"38ff73e804a2aa2bbd9eaa604154f67c21dd8dedb6da7fb068a0669b29ab6266"}}