{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:UEFCLG624N6K3ZFB42I677TI2H","short_pith_number":"pith:UEFCLG62","canonical_record":{"source":{"id":"2502.01652","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-30T21:04:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c47beac7bd91177451b84cd12085587a8cc94bf559d3b33867a86881893f0b68","abstract_canon_sha256":"90d5788ee3dd5ac3a4af7b292774e7242ccdd0c00a7e2357ac7046c619548a34"},"schema_version":"1.0"},"canonical_sha256":"a10a259bdae37cade4a1e691effe68d1e56a8fb4895332e06d088357912d3a74","source":{"kind":"arxiv","id":"2502.01652","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.01652","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"arxiv_version","alias_value":"2502.01652v1","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01652","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"pith_short_12","alias_value":"UEFCLG624N6K","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"pith_short_16","alias_value":"UEFCLG624N6K3ZFB","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"pith_short_8","alias_value":"UEFCLG62","created_at":"2026-07-05T10:09:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:UEFCLG624N6K3ZFB42I677TI2H","target":"record","payload":{"canonical_record":{"source":{"id":"2502.01652","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-30T21:04:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c47beac7bd91177451b84cd12085587a8cc94bf559d3b33867a86881893f0b68","abstract_canon_sha256":"90d5788ee3dd5ac3a4af7b292774e7242ccdd0c00a7e2357ac7046c619548a34"},"schema_version":"1.0"},"canonical_sha256":"a10a259bdae37cade4a1e691effe68d1e56a8fb4895332e06d088357912d3a74","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:09:17.070947Z","signature_b64":"nJP/fnoeQ/u8VXXFxXSeSaALnupPGO3Uyp2FrHTdn9jCLfipflkFdCZh0gzfXlmrW0JLnT0fOn8ySHmoA5fMCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a10a259bdae37cade4a1e691effe68d1e56a8fb4895332e06d088357912d3a74","last_reissued_at":"2026-07-05T10:09:17.070429Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:09:17.070429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.01652","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":"2026-07-05T10:09:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6TD04XmSduE1k61H6GK4ijdhF5K+ppSTtxmgc/KcEkXyfmPcv+43jgSEdMHdZf5XRHjpeiEzFbfbAyrp+EtpCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:15:03.007615Z"},"content_sha256":"7eb828bf8830dd6dadc6b5edd6a3ce3dd9d1b7ed51895a0526b8ff776d423a8d","schema_version":"1.0","event_id":"sha256:7eb828bf8830dd6dadc6b5edd6a3ce3dd9d1b7ed51895a0526b8ff776d423a8d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:UEFCLG624N6K3ZFB42I677TI2H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Hybrid Group Relative Policy Optimization: A Multi-Sample Approach to Enhancing Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Soham Sane","submitted_at":"2025-01-30T21:04:01Z","abstract_excerpt":"Hybrid Group Relative Policy Optimization (Hybrid GRPO) is a reinforcement learning framework that extends Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO) by incorporating empirical multi-sample action evaluation while preserving the stability of value function-based learning. Unlike DeepSeek GRPO, which eliminates the value function in favor of purely empirical reward estimation, Hybrid GRPO introduces a structured advantage computation method that balances empirical action sampling with bootstrapped value estimation. This approach enhances sample efficiency, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01652","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/2502.01652/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":"2026-07-05T10:09:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"woJgN5xLVqcg3ExIs/AYkPIvhicky6vPqAw/9o6NB8ZgIfeoXh8jy/Lie6LTdckQ3YKuANvdCw2r38oR3/R0DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:15:03.008591Z"},"content_sha256":"69533115bbaaa064ee80e22be490d3ecaacd67764d118d937fb6dada731891f5","schema_version":"1.0","event_id":"sha256:69533115bbaaa064ee80e22be490d3ecaacd67764d118d937fb6dada731891f5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UEFCLG624N6K3ZFB42I677TI2H/bundle.json","state_url":"https://pith.science/pith/UEFCLG624N6K3ZFB42I677TI2H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UEFCLG624N6K3ZFB42I677TI2H/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":"2026-08-09T14:15:03Z","links":{"resolver":"https://pith.science/pith/UEFCLG624N6K3ZFB42I677TI2H","bundle":"https://pith.science/pith/UEFCLG624N6K3ZFB42I677TI2H/bundle.json","state":"https://pith.science/pith/UEFCLG624N6K3ZFB42I677TI2H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UEFCLG624N6K3ZFB42I677TI2H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UEFCLG624N6K3ZFB42I677TI2H","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":"90d5788ee3dd5ac3a4af7b292774e7242ccdd0c00a7e2357ac7046c619548a34","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-30T21:04:01Z","title_canon_sha256":"c47beac7bd91177451b84cd12085587a8cc94bf559d3b33867a86881893f0b68"},"schema_version":"1.0","source":{"id":"2502.01652","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.01652","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"arxiv_version","alias_value":"2502.01652v1","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01652","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"pith_short_12","alias_value":"UEFCLG624N6K","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"pith_short_16","alias_value":"UEFCLG624N6K3ZFB","created_at":"2026-07-05T10:09:17Z"},{"alias_kind":"pith_short_8","alias_value":"UEFCLG62","created_at":"2026-07-05T10:09:17Z"}],"graph_snapshots":[{"event_id":"sha256:69533115bbaaa064ee80e22be490d3ecaacd67764d118d937fb6dada731891f5","target":"graph","created_at":"2026-07-05T10:09:17Z","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/2502.01652/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Hybrid Group Relative Policy Optimization (Hybrid GRPO) is a reinforcement learning framework that extends Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO) by incorporating empirical multi-sample action evaluation while preserving the stability of value function-based learning. Unlike DeepSeek GRPO, which eliminates the value function in favor of purely empirical reward estimation, Hybrid GRPO introduces a structured advantage computation method that balances empirical action sampling with bootstrapped value estimation. This approach enhances sample efficiency, ","authors_text":"Soham Sane","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-30T21:04:01Z","title":"Hybrid Group Relative Policy Optimization: A Multi-Sample Approach to Enhancing Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01652","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":"sha256:7eb828bf8830dd6dadc6b5edd6a3ce3dd9d1b7ed51895a0526b8ff776d423a8d","target":"record","created_at":"2026-07-05T10:09:17Z","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":"90d5788ee3dd5ac3a4af7b292774e7242ccdd0c00a7e2357ac7046c619548a34","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-30T21:04:01Z","title_canon_sha256":"c47beac7bd91177451b84cd12085587a8cc94bf559d3b33867a86881893f0b68"},"schema_version":"1.0","source":{"id":"2502.01652","kind":"arxiv","version":1}},"canonical_sha256":"a10a259bdae37cade4a1e691effe68d1e56a8fb4895332e06d088357912d3a74","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a10a259bdae37cade4a1e691effe68d1e56a8fb4895332e06d088357912d3a74","first_computed_at":"2026-07-05T10:09:17.070429Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:09:17.070429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nJP/fnoeQ/u8VXXFxXSeSaALnupPGO3Uyp2FrHTdn9jCLfipflkFdCZh0gzfXlmrW0JLnT0fOn8ySHmoA5fMCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:09:17.070947Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.01652","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7eb828bf8830dd6dadc6b5edd6a3ce3dd9d1b7ed51895a0526b8ff776d423a8d","sha256:69533115bbaaa064ee80e22be490d3ecaacd67764d118d937fb6dada731891f5"],"state_sha256":"3deb80055dfe170e044c7b8eb1dd8d24e2eb4d85b4e5f663fd444f0e1243cacd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VHUy/pdpCtek2r6nI0QQwDGh5aUdMmhYjRRdLY6EOpV1OZom4xU662+PoAj43z5OLVo1Uoz7EkY25wku8tXqDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T14:15:03.015607Z","bundle_sha256":"5c495d639f7e377eeeabed6ac55abadbb662985112339455f6b42f7073d79d88"}}