{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:DQZWRNYPCJ5QDMQ7JFCT26DBZP","short_pith_number":"pith:DQZWRNYP","canonical_record":{"source":{"id":"2506.11902","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-13T15:52:37Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2a747faa4fc974c86a1a7841837e6959c70e2f1cd2eef4cfe9dcf9e35a9d85f5","abstract_canon_sha256":"73e7f1fbc75196d42008fbe7f7f4d917ff6fe75ca82408a7b7fb58000ca8284f"},"schema_version":"1.0"},"canonical_sha256":"1c3368b70f127b01b21f49453d7861cbe88a3bddfbb340ecc3806e2a9280ce42","source":{"kind":"arxiv","id":"2506.11902","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.11902","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"arxiv_version","alias_value":"2506.11902v1","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11902","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"pith_short_12","alias_value":"DQZWRNYPCJ5Q","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"pith_short_16","alias_value":"DQZWRNYPCJ5QDMQ7","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"pith_short_8","alias_value":"DQZWRNYP","created_at":"2026-07-05T11:21:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:DQZWRNYPCJ5QDMQ7JFCT26DBZP","target":"record","payload":{"canonical_record":{"source":{"id":"2506.11902","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-13T15:52:37Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2a747faa4fc974c86a1a7841837e6959c70e2f1cd2eef4cfe9dcf9e35a9d85f5","abstract_canon_sha256":"73e7f1fbc75196d42008fbe7f7f4d917ff6fe75ca82408a7b7fb58000ca8284f"},"schema_version":"1.0"},"canonical_sha256":"1c3368b70f127b01b21f49453d7861cbe88a3bddfbb340ecc3806e2a9280ce42","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:12.584784Z","signature_b64":"VUvL8WuWzgBvAuCG8svcONU+h8ImPFp6WvNMpPrSG7HOV7modk9Qi7jzDjQcs9dC8jLPtO9rr9jMwDRvEFDvDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1c3368b70f127b01b21f49453d7861cbe88a3bddfbb340ecc3806e2a9280ce42","last_reissued_at":"2026-07-05T11:21:12.584308Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:12.584308Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.11902","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":"2026-07-05T11:21:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zX60wC/EQG7RjBTEm+hdt9M7oiBW88081Kt0KssnA/8UArPbHCRhpWWyxUhlnO4x8CbmVar5gu3+HU5gFIBvBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-15T22:45:28.803395Z"},"content_sha256":"f690b5fb9e27e49dcb203773e0da9a493fb337b2c829cbc56ddf375553a40b34","schema_version":"1.0","event_id":"sha256:f690b5fb9e27e49dcb203773e0da9a493fb337b2c829cbc56ddf375553a40b34"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:DQZWRNYPCJ5QDMQ7JFCT26DBZP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TreeRL: LLM Reinforcement Learning with On-Policy Tree Search","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Jie Tang, Rui Lu, Yujiang Li, Yuxiao Dong, Zhenyu Hou, Ziniu Hu","submitted_at":"2025-06-13T15:52:37Z","abstract_excerpt":"Reinforcement learning (RL) with tree search has demonstrated superior performance in traditional reasoning tasks. Compared to conventional independent chain sampling strategies with outcome supervision, tree search enables better exploration of the reasoning space and provides dense, on-policy process rewards during RL training but remains under-explored in On-Policy LLM RL. We propose TreeRL, a reinforcement learning framework that directly incorporates on-policy tree search for RL training. Our approach includes intermediate supervision and eliminates the need for a separate reward model tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11902","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/2506.11902/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":"2026-07-05T11:21:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mgcpgwMLFZ68MrbQXJRoy1jpOdmmb6F+Ypu3/Bydo6BpwDSdXb5OycLUR48cpKdlwrza7+LDcDPf8NrXZD22Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-15T22:45:28.803753Z"},"content_sha256":"73ffc5205264150366c72d0b3b066b1a828170c7433243c06411fa5570dd97fc","schema_version":"1.0","event_id":"sha256:73ffc5205264150366c72d0b3b066b1a828170c7433243c06411fa5570dd97fc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP/bundle.json","state_url":"https://pith.science/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP/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":"2026-07-15T22:45:28Z","links":{"resolver":"https://pith.science/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP","bundle":"https://pith.science/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP/bundle.json","state":"https://pith.science/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DQZWRNYPCJ5QDMQ7JFCT26DBZP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DQZWRNYPCJ5QDMQ7JFCT26DBZP","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":"73e7f1fbc75196d42008fbe7f7f4d917ff6fe75ca82408a7b7fb58000ca8284f","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-13T15:52:37Z","title_canon_sha256":"2a747faa4fc974c86a1a7841837e6959c70e2f1cd2eef4cfe9dcf9e35a9d85f5"},"schema_version":"1.0","source":{"id":"2506.11902","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.11902","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"arxiv_version","alias_value":"2506.11902v1","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11902","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"pith_short_12","alias_value":"DQZWRNYPCJ5Q","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"pith_short_16","alias_value":"DQZWRNYPCJ5QDMQ7","created_at":"2026-07-05T11:21:12Z"},{"alias_kind":"pith_short_8","alias_value":"DQZWRNYP","created_at":"2026-07-05T11:21:12Z"}],"graph_snapshots":[{"event_id":"sha256:73ffc5205264150366c72d0b3b066b1a828170c7433243c06411fa5570dd97fc","target":"graph","created_at":"2026-07-05T11:21:12Z","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/2506.11902/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) with tree search has demonstrated superior performance in traditional reasoning tasks. Compared to conventional independent chain sampling strategies with outcome supervision, tree search enables better exploration of the reasoning space and provides dense, on-policy process rewards during RL training but remains under-explored in On-Policy LLM RL. We propose TreeRL, a reinforcement learning framework that directly incorporates on-policy tree search for RL training. Our approach includes intermediate supervision and eliminates the need for a separate reward model tr","authors_text":"Jie Tang, Rui Lu, Yujiang Li, Yuxiao Dong, Zhenyu Hou, Ziniu Hu","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-13T15:52:37Z","title":"TreeRL: LLM Reinforcement Learning with On-Policy Tree Search"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11902","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:f690b5fb9e27e49dcb203773e0da9a493fb337b2c829cbc56ddf375553a40b34","target":"record","created_at":"2026-07-05T11:21:12Z","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":"73e7f1fbc75196d42008fbe7f7f4d917ff6fe75ca82408a7b7fb58000ca8284f","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-13T15:52:37Z","title_canon_sha256":"2a747faa4fc974c86a1a7841837e6959c70e2f1cd2eef4cfe9dcf9e35a9d85f5"},"schema_version":"1.0","source":{"id":"2506.11902","kind":"arxiv","version":1}},"canonical_sha256":"1c3368b70f127b01b21f49453d7861cbe88a3bddfbb340ecc3806e2a9280ce42","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1c3368b70f127b01b21f49453d7861cbe88a3bddfbb340ecc3806e2a9280ce42","first_computed_at":"2026-07-05T11:21:12.584308Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:21:12.584308Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VUvL8WuWzgBvAuCG8svcONU+h8ImPFp6WvNMpPrSG7HOV7modk9Qi7jzDjQcs9dC8jLPtO9rr9jMwDRvEFDvDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:21:12.584784Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.11902","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f690b5fb9e27e49dcb203773e0da9a493fb337b2c829cbc56ddf375553a40b34","sha256:73ffc5205264150366c72d0b3b066b1a828170c7433243c06411fa5570dd97fc"],"state_sha256":"77e66eae21129aedec61dcb1bc2423473730a1572fdfa66e41322e8268241627"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MCZEgSuOJdKPsu4+LtxTe0ZKsplkLD+jaTNW2UFvmuucDrMgR0heTRnX6DudOg2IP0OB8rhlNmVF3IJK4LGIDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-15T22:45:28.805864Z","bundle_sha256":"eec66948cac1c68ca6334a5352b6b25ce9a24dad8c14d0a60b5be671872eeb53"}}