{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:VGWMCXOXY7FOFEOELU7FBFGGUM","short_pith_number":"pith:VGWMCXOX","canonical_record":{"source":{"id":"2510.26270","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-10-30T08:53:41Z","cross_cats_sorted":[],"title_canon_sha256":"3dc77d301413ff029929fd57e44047e1a49b6186e1a39c6844defad82424e48b","abstract_canon_sha256":"12fb5d7cda3346a09ac24f0f7a1a96e2f129842a7dccec3d3d62ad38fcfab744"},"schema_version":"1.0"},"canonical_sha256":"a9acc15dd7c7cae291c45d3e5094c6a318496f9c03fd20236246f9c1a045c983","source":{"kind":"arxiv","id":"2510.26270","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.26270","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"arxiv_version","alias_value":"2510.26270v2","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.26270","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"pith_short_12","alias_value":"VGWMCXOXY7FO","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"pith_short_16","alias_value":"VGWMCXOXY7FOFEOE","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"pith_short_8","alias_value":"VGWMCXOX","created_at":"2026-05-29T01:04:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:VGWMCXOXY7FOFEOELU7FBFGGUM","target":"record","payload":{"canonical_record":{"source":{"id":"2510.26270","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-10-30T08:53:41Z","cross_cats_sorted":[],"title_canon_sha256":"3dc77d301413ff029929fd57e44047e1a49b6186e1a39c6844defad82424e48b","abstract_canon_sha256":"12fb5d7cda3346a09ac24f0f7a1a96e2f129842a7dccec3d3d62ad38fcfab744"},"schema_version":"1.0"},"canonical_sha256":"a9acc15dd7c7cae291c45d3e5094c6a318496f9c03fd20236246f9c1a045c983","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:04:57.560030Z","signature_b64":"ggZYj16PfSvkNiDfjnXRAtmzhEH3tFS3K52SmW3TMZger2bmEiNh/BuMR9oxXQtmFCb9sTT4fwTlWSt6VQx4CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9acc15dd7c7cae291c45d3e5094c6a318496f9c03fd20236246f9c1a045c983","last_reissued_at":"2026-05-29T01:04:57.559509Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:04:57.559509Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2510.26270","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":"2026-05-29T01:04:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MrfNGqfGUSE3YXSBXWjPapEy6Uwtpeb8yuQeZ15Gofhq5x+czBYGvlBeQJ48Pzt/WstrEy0xuJ/jr+kI0nCTAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T20:50:15.011172Z"},"content_sha256":"3951298f3dbd45f0d67085df7b182d31cfc9eb68cf67eb6737bd3997b6c4e0e4","schema_version":"1.0","event_id":"sha256:3951298f3dbd45f0d67085df7b182d31cfc9eb68cf67eb6737bd3997b6c4e0e4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:VGWMCXOXY7FOFEOELU7FBFGGUM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Graph-Enhanced Policy Optimization in LLM Agent Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jiazhen Yuan, Jinquan Hang, Wei Zhao, Zhengbiao Bai, Zhike Gong","submitted_at":"2025-10-30T08:53:41Z","abstract_excerpt":"Multi-step LLM agents in interactive environments represent a crucial step toward long-horizon decision-making. To train such agents, group-based reinforcement learning is widely adopted, which reinforces trajectories with higher relative performance within the group. However, in most existing methods, every step within a trajectory and every trajectory with the same terminal reward receive identical credit, regardless of their actual contributions. Since different states play different structural roles in an online state-transition graph built from sampled trajectories, their impacts should b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.26270","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/2510.26270/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":"2026-05-29T01:04:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wJFM+jFU4McrVTRASLeUcxlz3QlNLWwp+vwwwZO4zuMDARanvWR9Wu1DhJotQ4pB6UEArWr0ppIHd0Kh8/fPBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T20:50:15.011572Z"},"content_sha256":"a97804398e8c86adc74b716266ba992a6109f9afdb2974f9952eab1f53564def","schema_version":"1.0","event_id":"sha256:a97804398e8c86adc74b716266ba992a6109f9afdb2974f9952eab1f53564def"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VGWMCXOXY7FOFEOELU7FBFGGUM/bundle.json","state_url":"https://pith.science/pith/VGWMCXOXY7FOFEOELU7FBFGGUM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VGWMCXOXY7FOFEOELU7FBFGGUM/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":"2026-06-10T20:50:15Z","links":{"resolver":"https://pith.science/pith/VGWMCXOXY7FOFEOELU7FBFGGUM","bundle":"https://pith.science/pith/VGWMCXOXY7FOFEOELU7FBFGGUM/bundle.json","state":"https://pith.science/pith/VGWMCXOXY7FOFEOELU7FBFGGUM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VGWMCXOXY7FOFEOELU7FBFGGUM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VGWMCXOXY7FOFEOELU7FBFGGUM","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":"12fb5d7cda3346a09ac24f0f7a1a96e2f129842a7dccec3d3d62ad38fcfab744","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-10-30T08:53:41Z","title_canon_sha256":"3dc77d301413ff029929fd57e44047e1a49b6186e1a39c6844defad82424e48b"},"schema_version":"1.0","source":{"id":"2510.26270","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.26270","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"arxiv_version","alias_value":"2510.26270v2","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.26270","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"pith_short_12","alias_value":"VGWMCXOXY7FO","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"pith_short_16","alias_value":"VGWMCXOXY7FOFEOE","created_at":"2026-05-29T01:04:57Z"},{"alias_kind":"pith_short_8","alias_value":"VGWMCXOX","created_at":"2026-05-29T01:04:57Z"}],"graph_snapshots":[{"event_id":"sha256:a97804398e8c86adc74b716266ba992a6109f9afdb2974f9952eab1f53564def","target":"graph","created_at":"2026-05-29T01:04:57Z","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/2510.26270/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-step LLM agents in interactive environments represent a crucial step toward long-horizon decision-making. To train such agents, group-based reinforcement learning is widely adopted, which reinforces trajectories with higher relative performance within the group. However, in most existing methods, every step within a trajectory and every trajectory with the same terminal reward receive identical credit, regardless of their actual contributions. Since different states play different structural roles in an online state-transition graph built from sampled trajectories, their impacts should b","authors_text":"Jiazhen Yuan, Jinquan Hang, Wei Zhao, Zhengbiao Bai, Zhike Gong","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-10-30T08:53:41Z","title":"Graph-Enhanced Policy Optimization in LLM Agent Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.26270","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":"sha256:3951298f3dbd45f0d67085df7b182d31cfc9eb68cf67eb6737bd3997b6c4e0e4","target":"record","created_at":"2026-05-29T01:04:57Z","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":"12fb5d7cda3346a09ac24f0f7a1a96e2f129842a7dccec3d3d62ad38fcfab744","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-10-30T08:53:41Z","title_canon_sha256":"3dc77d301413ff029929fd57e44047e1a49b6186e1a39c6844defad82424e48b"},"schema_version":"1.0","source":{"id":"2510.26270","kind":"arxiv","version":2}},"canonical_sha256":"a9acc15dd7c7cae291c45d3e5094c6a318496f9c03fd20236246f9c1a045c983","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a9acc15dd7c7cae291c45d3e5094c6a318496f9c03fd20236246f9c1a045c983","first_computed_at":"2026-05-29T01:04:57.559509Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:04:57.559509Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ggZYj16PfSvkNiDfjnXRAtmzhEH3tFS3K52SmW3TMZger2bmEiNh/BuMR9oxXQtmFCb9sTT4fwTlWSt6VQx4CQ==","signature_status":"signed_v1","signed_at":"2026-05-29T01:04:57.560030Z","signed_message":"canonical_sha256_bytes"},"source_id":"2510.26270","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3951298f3dbd45f0d67085df7b182d31cfc9eb68cf67eb6737bd3997b6c4e0e4","sha256:a97804398e8c86adc74b716266ba992a6109f9afdb2974f9952eab1f53564def"],"state_sha256":"7a4eba9eb4018ac69782c4eaa2c649a446d7e7eda16d9b0a8191dd46e28b1e77"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"awJ+JzQ12uazl+/nWJSFMO6P0RlYseJxxWEdiEbeZpobGsrbSGpr4RtHWM7iu83Yt2AuH0bhSzd835G8X8gaCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T20:50:15.013609Z","bundle_sha256":"4e1643e9e6b45eef554ba971e06536ac142ee35590074768d1a4f7ff95b5a0f2"}}