{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GYUUNYSLU4Z7ZZYCIGS44JHN6Q","short_pith_number":"pith:GYUUNYSL","schema_version":"1.0","canonical_sha256":"362946e24ba733fce70241a5ce24edf42e2dda56b224a825129bdc38df876dce","source":{"kind":"arxiv","id":"2606.22995","version":1},"attestation_state":"computed","paper":{"title":"Group-Graph Policy Optimization for Long-Horizon Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Feng Sun, Furu Wei, Haizhen Huang, MingHui Song, Qi Zhang, Shaohan Huang, Weiwei Deng, Yunan Wang, Zihan Zhang","submitted_at":"2026-06-22T08:12:47Z","abstract_excerpt":"Group-based Reinforcement Learning (RL) has significantly enhanced Large Language Models (LLMs) in agentic scenarios. To achieve finer-grained policy updates, recent agentic RL frameworks have shifted from trajectory-level to step-level training. However, long-horizon agentic RL suffers from severe reward sparsity and delay, as feedback is often deferred for dozens of interaction steps. While existing step-level frameworks refine training granularity, their credit assignment remains coarse-grained and still treats agent exploration as isolated, linear trajectories. This oversimplified perspect"},"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":"2606.22995","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-22T08:12:47Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"89ea7d9958648adb41e0bba207e399edeea8255f0f5bee5ecfd54b0a1269c645","abstract_canon_sha256":"b8e058b486fc86f630ffdaa63f134025635a1c0209aaa99df507687de2e64465"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:14:06.232055Z","signature_b64":"6lZvwiN3v6CJ+2InTegpBRs/d4MImel2cGBkbjg50kuaaYUUa4+EJq8gsoqaKnid91YHWpBwOFQEGrmO5BbTDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"362946e24ba733fce70241a5ce24edf42e2dda56b224a825129bdc38df876dce","last_reissued_at":"2026-06-23T03:14:06.231716Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:14:06.231716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Group-Graph Policy Optimization for Long-Horizon Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Feng Sun, Furu Wei, Haizhen Huang, MingHui Song, Qi Zhang, Shaohan Huang, Weiwei Deng, Yunan Wang, Zihan Zhang","submitted_at":"2026-06-22T08:12:47Z","abstract_excerpt":"Group-based Reinforcement Learning (RL) has significantly enhanced Large Language Models (LLMs) in agentic scenarios. To achieve finer-grained policy updates, recent agentic RL frameworks have shifted from trajectory-level to step-level training. However, long-horizon agentic RL suffers from severe reward sparsity and delay, as feedback is often deferred for dozens of interaction steps. While existing step-level frameworks refine training granularity, their credit assignment remains coarse-grained and still treats agent exploration as isolated, linear trajectories. This oversimplified perspect"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22995","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/2606.22995/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":"2606.22995","created_at":"2026-06-23T03:14:06.231773+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22995v1","created_at":"2026-06-23T03:14:06.231773+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22995","created_at":"2026-06-23T03:14:06.231773+00:00"},{"alias_kind":"pith_short_12","alias_value":"GYUUNYSLU4Z7","created_at":"2026-06-23T03:14:06.231773+00:00"},{"alias_kind":"pith_short_16","alias_value":"GYUUNYSLU4Z7ZZYC","created_at":"2026-06-23T03:14:06.231773+00:00"},{"alias_kind":"pith_short_8","alias_value":"GYUUNYSL","created_at":"2026-06-23T03:14:06.231773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q","json":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q.json","graph_json":"https://pith.science/api/pith-number/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/graph.json","events_json":"https://pith.science/api/pith-number/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/events.json","paper":"https://pith.science/paper/GYUUNYSL"},"agent_actions":{"view_html":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q","download_json":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q.json","view_paper":"https://pith.science/paper/GYUUNYSL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22995&json=true","fetch_graph":"https://pith.science/api/pith-number/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/graph.json","fetch_events":"https://pith.science/api/pith-number/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/action/storage_attestation","attest_author":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/action/author_attestation","sign_citation":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/action/citation_signature","submit_replication":"https://pith.science/pith/GYUUNYSLU4Z7ZZYCIGS44JHN6Q/action/replication_record"}},"created_at":"2026-06-23T03:14:06.231773+00:00","updated_at":"2026-06-23T03:14:06.231773+00:00"}