{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:GBMGNK6Q6ZHYSDHGILMEQ2UDMC","short_pith_number":"pith:GBMGNK6Q","canonical_record":{"source":{"id":"2604.18401","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-20T15:22:39Z","cross_cats_sorted":[],"title_canon_sha256":"9b0768938686a95fe0912f58ee7dfc52476fb36ae3bc01c12c7a51db13f1f769","abstract_canon_sha256":"d92a6bb55826794aaea03e8bdd224d72c737bc0a610264b10ccbaf710c99a3c9"},"schema_version":"1.0"},"canonical_sha256":"305866abd0f64f890ce642d8486a8360a67d518fef2ed4d01d29bb17a1bed39f","source":{"kind":"arxiv","id":"2604.18401","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.18401","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"arxiv_version","alias_value":"2604.18401v2","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.18401","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"pith_short_12","alias_value":"GBMGNK6Q6ZHY","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"pith_short_16","alias_value":"GBMGNK6Q6ZHYSDHG","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"pith_short_8","alias_value":"GBMGNK6Q","created_at":"2026-06-02T02:04:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:GBMGNK6Q6ZHYSDHGILMEQ2UDMC","target":"record","payload":{"canonical_record":{"source":{"id":"2604.18401","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-20T15:22:39Z","cross_cats_sorted":[],"title_canon_sha256":"9b0768938686a95fe0912f58ee7dfc52476fb36ae3bc01c12c7a51db13f1f769","abstract_canon_sha256":"d92a6bb55826794aaea03e8bdd224d72c737bc0a610264b10ccbaf710c99a3c9"},"schema_version":"1.0"},"canonical_sha256":"305866abd0f64f890ce642d8486a8360a67d518fef2ed4d01d29bb17a1bed39f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:53.338536Z","signature_b64":"yP2Nsh2+YOecJiM6Sgddi8eJaX1kpiPmAf3X/NE/HptA/41AYPdSz51awUFlMMzJSiI/5kdfbRtdJetnn96ZBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"305866abd0f64f890ce642d8486a8360a67d518fef2ed4d01d29bb17a1bed39f","last_reissued_at":"2026-06-02T02:04:53.338134Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:53.338134Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.18401","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":"2026-06-02T02:04:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z3cqRDS+nlRxlEZJdSggsSZ7NM8mAxZ9wyA+UIwWcWnvWYJHtTvHkKto993hEyPbFL5rSSUBdWieyLP+fmSSDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T20:31:58.387347Z"},"content_sha256":"9f46ec284695587367947198d5a66dd591feb7e6566293249c3c1c177b5988a3","schema_version":"1.0","event_id":"sha256:9f46ec284695587367947198d5a66dd591feb7e6566293249c3c1c177b5988a3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:GBMGNK6Q6ZHYSDHGILMEQ2UDMC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"StepPO: Step-Aligned Policy Optimization for Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LLM agents need step-level MDP and credit assignment rather than token-level modeling for multi-turn RL.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Daoyu Wang, Enhong Chen, Jie Ouyang, Mingyue Cheng, Qi Liu, Qingchuan Li, Shuo Yu","submitted_at":"2026-04-20T15:22:39Z","abstract_excerpt":"Agentic reinforcement learning (RL) is emerging as a critical post-training paradigm for improving LLM agent capabilities. Existing RL algorithms for LLMs largely follow the token-centric paradigm as in RLHF and RLVR, where tokens serve as the basic units for modeling and optimization. However, this paradigm introduces a granularity mismatch in agentic RL, as it optimizes token-level predictions while LLM agents make step-level decisions through cycles of environmental observations and actions. To bridge this gap, we propose \\textbf{StepPO}, a step-centric paradigm for agentic RL via step-alig"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"the conventional token-level Markov Decision Process (MDP) should be advanced to a step-level MDP formulation, and that the step, rather than the token, should be regarded as the proper action representation for LLM agents","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That redefining the MDP and credit assignment at step granularity will meaningfully address delayed/sparse rewards and long context challenges in multi-turn agent settings.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"StepPO argues that LLM agents should optimize at the step level rather than token level to better handle delayed rewards and long contexts in agentic RL.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM agents need step-level MDP and credit assignment rather than token-level modeling for multi-turn RL.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a13be7d3ed31acb1022e48c7c1388a19ea717b0a6749fc0fe81cb0f2cbaefa1d"},"source":{"id":"2604.18401","kind":"arxiv","version":2},"verdict":{"id":"9f7497bf-1d29-4123-a5db-df28933e1a61","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T04:23:57.354913Z","strongest_claim":"the conventional token-level Markov Decision Process (MDP) should be advanced to a step-level MDP formulation, and that the step, rather than the token, should be regarded as the proper action representation for LLM agents","one_line_summary":"StepPO argues that LLM agents should optimize at the step level rather than token level to better handle delayed rewards and long contexts in agentic RL.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That redefining the MDP and credit assignment at step granularity will meaningfully address delayed/sparse rewards and long context challenges in multi-turn agent settings.","pith_extraction_headline":"LLM agents need step-level MDP and credit assignment rather than token-level modeling for multi-turn RL."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.18401/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-20T04:02:34.505902Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"e916b5c1865fb0d190821c203dfeac2c3e81eb7a3f5b3bc71143c18d8b737b04"},"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":"9f7497bf-1d29-4123-a5db-df28933e1a61"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oDuHgYgoX3CnXhNnl4CebMFbF3Ag7Qt73CQB29iXJ+ms0P7gL8DHhT49mdQqeXa0BQu/2QORXV5EV9odhCuiAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T20:31:58.387801Z"},"content_sha256":"faa5bea6619c1b7be7b6effa563fffc9c034b59b1a8207afb7a59289f5fc5519","schema_version":"1.0","event_id":"sha256:faa5bea6619c1b7be7b6effa563fffc9c034b59b1a8207afb7a59289f5fc5519"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC/bundle.json","state_url":"https://pith.science/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC/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":"2026-06-28T20:31:58Z","links":{"resolver":"https://pith.science/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC","bundle":"https://pith.science/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC/bundle.json","state":"https://pith.science/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GBMGNK6Q6ZHYSDHGILMEQ2UDMC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GBMGNK6Q6ZHYSDHGILMEQ2UDMC","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":"d92a6bb55826794aaea03e8bdd224d72c737bc0a610264b10ccbaf710c99a3c9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-20T15:22:39Z","title_canon_sha256":"9b0768938686a95fe0912f58ee7dfc52476fb36ae3bc01c12c7a51db13f1f769"},"schema_version":"1.0","source":{"id":"2604.18401","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.18401","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"arxiv_version","alias_value":"2604.18401v2","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.18401","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"pith_short_12","alias_value":"GBMGNK6Q6ZHY","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"pith_short_16","alias_value":"GBMGNK6Q6ZHYSDHG","created_at":"2026-06-02T02:04:53Z"},{"alias_kind":"pith_short_8","alias_value":"GBMGNK6Q","created_at":"2026-06-02T02:04:53Z"}],"graph_snapshots":[{"event_id":"sha256:faa5bea6619c1b7be7b6effa563fffc9c034b59b1a8207afb7a59289f5fc5519","target":"graph","created_at":"2026-06-02T02:04:53Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"the conventional token-level Markov Decision Process (MDP) should be advanced to a step-level MDP formulation, and that the step, rather than the token, should be regarded as the proper action representation for LLM agents"},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That redefining the MDP and credit assignment at step granularity will meaningfully address delayed/sparse rewards and long context challenges in multi-turn agent settings."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"StepPO argues that LLM agents should optimize at the step level rather than token level to better handle delayed rewards and long contexts in agentic RL."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLM agents need step-level MDP and credit assignment rather than token-level modeling for multi-turn RL."}],"snapshot_sha256":"a13be7d3ed31acb1022e48c7c1388a19ea717b0a6749fc0fe81cb0f2cbaefa1d"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-20T04:02:34.505902Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.18401/integrity.json","findings":[],"snapshot_sha256":"e916b5c1865fb0d190821c203dfeac2c3e81eb7a3f5b3bc71143c18d8b737b04","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Agentic reinforcement learning (RL) is emerging as a critical post-training paradigm for improving LLM agent capabilities. Existing RL algorithms for LLMs largely follow the token-centric paradigm as in RLHF and RLVR, where tokens serve as the basic units for modeling and optimization. However, this paradigm introduces a granularity mismatch in agentic RL, as it optimizes token-level predictions while LLM agents make step-level decisions through cycles of environmental observations and actions. To bridge this gap, we propose \\textbf{StepPO}, a step-centric paradigm for agentic RL via step-alig","authors_text":"Daoyu Wang, Enhong Chen, Jie Ouyang, Mingyue Cheng, Qi Liu, Qingchuan Li, Shuo Yu","cross_cats":[],"headline":"LLM agents need step-level MDP and credit assignment rather than token-level modeling for multi-turn RL.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-20T15:22:39Z","title":"StepPO: Step-Aligned Policy Optimization for Agentic Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.18401","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T04:23:57.354913Z","id":"9f7497bf-1d29-4123-a5db-df28933e1a61","model_set":{"reader":"grok-4.3"},"one_line_summary":"StepPO argues that LLM agents should optimize at the step level rather than token level to better handle delayed rewards and long contexts in agentic RL.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"LLM agents need step-level MDP and credit assignment rather than token-level modeling for multi-turn RL.","strongest_claim":"the conventional token-level Markov Decision Process (MDP) should be advanced to a step-level MDP formulation, and that the step, rather than the token, should be regarded as the proper action representation for LLM agents","weakest_assumption":"That redefining the MDP and credit assignment at step granularity will meaningfully address delayed/sparse rewards and long context challenges in multi-turn agent settings."}},"verdict_id":"9f7497bf-1d29-4123-a5db-df28933e1a61"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9f46ec284695587367947198d5a66dd591feb7e6566293249c3c1c177b5988a3","target":"record","created_at":"2026-06-02T02:04:53Z","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":"d92a6bb55826794aaea03e8bdd224d72c737bc0a610264b10ccbaf710c99a3c9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-20T15:22:39Z","title_canon_sha256":"9b0768938686a95fe0912f58ee7dfc52476fb36ae3bc01c12c7a51db13f1f769"},"schema_version":"1.0","source":{"id":"2604.18401","kind":"arxiv","version":2}},"canonical_sha256":"305866abd0f64f890ce642d8486a8360a67d518fef2ed4d01d29bb17a1bed39f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"305866abd0f64f890ce642d8486a8360a67d518fef2ed4d01d29bb17a1bed39f","first_computed_at":"2026-06-02T02:04:53.338134Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:53.338134Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yP2Nsh2+YOecJiM6Sgddi8eJaX1kpiPmAf3X/NE/HptA/41AYPdSz51awUFlMMzJSiI/5kdfbRtdJetnn96ZBQ==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:53.338536Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.18401","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9f46ec284695587367947198d5a66dd591feb7e6566293249c3c1c177b5988a3","sha256:faa5bea6619c1b7be7b6effa563fffc9c034b59b1a8207afb7a59289f5fc5519"],"state_sha256":"68e2814d6d781c36918ea53dc363588e4b712285a381423071e3bd65bcd01d41"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T0l96BasKoDI5bWYWrzKpnVQD1w8ohjhLNn2vkp5GAFuWMGr3J/mJ8gBv6Ru6SrDMxtR33THZM0gVxVKOv0cAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T20:31:58.389953Z","bundle_sha256":"0830a157033695c8cc107f54d6605e36c9698b6804ad31f3e0dfb075cd37529f"}}