{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NI66TQHPRGNOVKCWLBAKR4FWOP","short_pith_number":"pith:NI66TQHP","canonical_record":{"source":{"id":"2603.18859","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-19T13:07:12Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"61c8af3615987a12c6b9f90cea4712d04b3b83c21fa58622bde89b18fe4281be","abstract_canon_sha256":"1677d4b04f23b0e67484718948d75c1ee88136613ecb5f104bb8704b26623812"},"schema_version":"1.0"},"canonical_sha256":"6a3de9c0ef899aeaa8565840a8f0b673f192c1f9a2d3cf1b7e4367955958accf","source":{"kind":"arxiv","id":"2603.18859","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.18859","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"arxiv_version","alias_value":"2603.18859v2","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.18859","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"pith_short_12","alias_value":"NI66TQHPRGNO","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"pith_short_16","alias_value":"NI66TQHPRGNOVKCW","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"pith_short_8","alias_value":"NI66TQHP","created_at":"2026-05-29T01:05:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NI66TQHPRGNOVKCWLBAKR4FWOP","target":"record","payload":{"canonical_record":{"source":{"id":"2603.18859","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-19T13:07:12Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"61c8af3615987a12c6b9f90cea4712d04b3b83c21fa58622bde89b18fe4281be","abstract_canon_sha256":"1677d4b04f23b0e67484718948d75c1ee88136613ecb5f104bb8704b26623812"},"schema_version":"1.0"},"canonical_sha256":"6a3de9c0ef899aeaa8565840a8f0b673f192c1f9a2d3cf1b7e4367955958accf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:07.575887Z","signature_b64":"7U59NPeQgZs9egc0p18maPGzcbgfhakFoioiYCUwQ7LgZrTdQf59CGlRroCaL5daqak3u0+gdiysziHIZ+HiCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a3de9c0ef899aeaa8565840a8f0b673f192c1f9a2d3cf1b7e4367955958accf","last_reissued_at":"2026-05-29T01:05:07.574764Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:07.574764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.18859","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":"2026-05-29T01:05:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/DaX1HkpJFzjRrsjw1FWeN2+HJQXlKgEiiCr6q+LQBaXdBqQmo3hOa5wVNjjJl+Z9yKJd0Gm8eHg8MDKAAR2BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T15:29:05.561785Z"},"content_sha256":"d000ee8104956daaece5c7f1f04b0691173625f655e51dedee2bb5d6d6ec4819","schema_version":"1.0","event_id":"sha256:d000ee8104956daaece5c7f1f04b0691173625f655e51dedee2bb5d6d6ec4819"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NI66TQHPRGNOVKCWLBAKR4FWOP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RewardFlow: Topology-Aware Reward Propagation on State Graphs for Agentic RL with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Bo Han, Dahai Yu, Jiangchao Yao, Jiaqi Fan, Ka Ho Li, Michael Kwok-Po Ng, Xiao Feng, Zhanke Zhou","submitted_at":"2026-03-19T13:07:12Z","abstract_excerpt":"Reinforcement learning (RL) shows promise for enhancing LLM agentic reasoning, yet sparse terminal rewards hinder fine-grained optimization. Process reward modeling offers an alternative but incurs high computational costs, reward hacking risks, and annotation bottlenecks. We introduce RewardFlow, a lightweight method for estimating state-level rewards in agentic reasoning. By constructing state graphs that capture the intrinsic topological structure of trajectories, RewardFlow performs topology-aware propagation to estimate each state's contribution to success, yielding principled, annotation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.18859","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/2603.18859/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":"2026-05-29T01:05:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GVS6Vx+JU5Nmp5TNM90cYOUK20b15FwcohuUp1EnLwSQaSz5czRP9SM5Kr+w/YSpSe/JimXthqz1O+NZmfw9CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T15:29:05.562167Z"},"content_sha256":"41564578611aa56e5bdcae370f553fdc9388c86f066dc699ad45d5b5eb877c58","schema_version":"1.0","event_id":"sha256:41564578611aa56e5bdcae370f553fdc9388c86f066dc699ad45d5b5eb877c58"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NI66TQHPRGNOVKCWLBAKR4FWOP/bundle.json","state_url":"https://pith.science/pith/NI66TQHPRGNOVKCWLBAKR4FWOP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NI66TQHPRGNOVKCWLBAKR4FWOP/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":"2026-06-01T15:29:05Z","links":{"resolver":"https://pith.science/pith/NI66TQHPRGNOVKCWLBAKR4FWOP","bundle":"https://pith.science/pith/NI66TQHPRGNOVKCWLBAKR4FWOP/bundle.json","state":"https://pith.science/pith/NI66TQHPRGNOVKCWLBAKR4FWOP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NI66TQHPRGNOVKCWLBAKR4FWOP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NI66TQHPRGNOVKCWLBAKR4FWOP","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":"1677d4b04f23b0e67484718948d75c1ee88136613ecb5f104bb8704b26623812","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-19T13:07:12Z","title_canon_sha256":"61c8af3615987a12c6b9f90cea4712d04b3b83c21fa58622bde89b18fe4281be"},"schema_version":"1.0","source":{"id":"2603.18859","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.18859","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"arxiv_version","alias_value":"2603.18859v2","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.18859","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"pith_short_12","alias_value":"NI66TQHPRGNO","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"pith_short_16","alias_value":"NI66TQHPRGNOVKCW","created_at":"2026-05-29T01:05:07Z"},{"alias_kind":"pith_short_8","alias_value":"NI66TQHP","created_at":"2026-05-29T01:05:07Z"}],"graph_snapshots":[{"event_id":"sha256:41564578611aa56e5bdcae370f553fdc9388c86f066dc699ad45d5b5eb877c58","target":"graph","created_at":"2026-05-29T01:05:07Z","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/2603.18859/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) shows promise for enhancing LLM agentic reasoning, yet sparse terminal rewards hinder fine-grained optimization. Process reward modeling offers an alternative but incurs high computational costs, reward hacking risks, and annotation bottlenecks. We introduce RewardFlow, a lightweight method for estimating state-level rewards in agentic reasoning. By constructing state graphs that capture the intrinsic topological structure of trajectories, RewardFlow performs topology-aware propagation to estimate each state's contribution to success, yielding principled, annotation","authors_text":"Bo Han, Dahai Yu, Jiangchao Yao, Jiaqi Fan, Ka Ho Li, Michael Kwok-Po Ng, Xiao Feng, Zhanke Zhou","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-19T13:07:12Z","title":"RewardFlow: Topology-Aware Reward Propagation on State Graphs for Agentic RL with Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.18859","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:d000ee8104956daaece5c7f1f04b0691173625f655e51dedee2bb5d6d6ec4819","target":"record","created_at":"2026-05-29T01:05:07Z","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":"1677d4b04f23b0e67484718948d75c1ee88136613ecb5f104bb8704b26623812","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-19T13:07:12Z","title_canon_sha256":"61c8af3615987a12c6b9f90cea4712d04b3b83c21fa58622bde89b18fe4281be"},"schema_version":"1.0","source":{"id":"2603.18859","kind":"arxiv","version":2}},"canonical_sha256":"6a3de9c0ef899aeaa8565840a8f0b673f192c1f9a2d3cf1b7e4367955958accf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6a3de9c0ef899aeaa8565840a8f0b673f192c1f9a2d3cf1b7e4367955958accf","first_computed_at":"2026-05-29T01:05:07.574764Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:05:07.574764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7U59NPeQgZs9egc0p18maPGzcbgfhakFoioiYCUwQ7LgZrTdQf59CGlRroCaL5daqak3u0+gdiysziHIZ+HiCw==","signature_status":"signed_v1","signed_at":"2026-05-29T01:05:07.575887Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.18859","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d000ee8104956daaece5c7f1f04b0691173625f655e51dedee2bb5d6d6ec4819","sha256:41564578611aa56e5bdcae370f553fdc9388c86f066dc699ad45d5b5eb877c58"],"state_sha256":"8e8736cf66ba1f922978818998728d1e74825ff153f465cab88d23e80e7a79f1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Cd0rpJfTdDqFcTT7OkTMNagrhnWq66jyLEdGDmv0gZTZvtvRY0uQIiDGPnd5YLYDX2DsYEul7VSu84YFaPsiDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T15:29:05.564298Z","bundle_sha256":"e9e356816e5f36cb99260408f1a612262de77c0a3c8c5d73b7cf789ec9f12542"}}