{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:CU2IPJNYP4JID5VBP3QMLAMOZM","short_pith_number":"pith:CU2IPJNY","canonical_record":{"source":{"id":"2605.21557","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-20T13:46:22Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"539be7504c6a3b01f872918005e53a0f847c006e739455e7870b362a8c6ad34e","abstract_canon_sha256":"91d4d2d5322547fb031f3df2e193c64c01cf0d598dbb2929ec7f2a57a3aba1e7"},"schema_version":"1.0"},"canonical_sha256":"153487a5b87f1281f6a17ee0c5818ecb3f0a656cb52a2413c496bb47601acab7","source":{"kind":"arxiv","id":"2605.21557","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21557","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21557v1","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21557","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"pith_short_12","alias_value":"CU2IPJNYP4JI","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"pith_short_16","alias_value":"CU2IPJNYP4JID5VB","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"pith_short_8","alias_value":"CU2IPJNY","created_at":"2026-05-22T00:02:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:CU2IPJNYP4JID5VBP3QMLAMOZM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.21557","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-20T13:46:22Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"539be7504c6a3b01f872918005e53a0f847c006e739455e7870b362a8c6ad34e","abstract_canon_sha256":"91d4d2d5322547fb031f3df2e193c64c01cf0d598dbb2929ec7f2a57a3aba1e7"},"schema_version":"1.0"},"canonical_sha256":"153487a5b87f1281f6a17ee0c5818ecb3f0a656cb52a2413c496bb47601acab7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T00:02:28.410596Z","signature_b64":"/QdExPrIAmkKeo2z64N59hKr5CcfxTJY7UbWTVjEAXGRgpuLl9akk6n0BhrANvuSEXkeOXoj3LrMdK5rwns9Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"153487a5b87f1281f6a17ee0c5818ecb3f0a656cb52a2413c496bb47601acab7","last_reissued_at":"2026-05-22T00:02:28.410043Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T00:02:28.410043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.21557","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":"2026-05-22T00:02:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b4BTvyZW6ec3v+4cZOSGGgLCArRvFvjBgaLdcLQF1P8K0d6lwdZSA3d+YdMSlqErQWSZRHnnba8jf9COQMx9BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T15:23:50.804642Z"},"content_sha256":"0ca42d388553016acf1aaeda422b1c0afb64a7e0a81d600e6edf010f845dd2ca","schema_version":"1.0","event_id":"sha256:0ca42d388553016acf1aaeda422b1c0afb64a7e0a81d600e6edf010f845dd2ca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:CU2IPJNYP4JID5VBP3QMLAMOZM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scalable On-Policy Reinforcement Learning via Adaptive Batch Scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Jongchan Park","submitted_at":"2026-05-20T13:46:22Z","abstract_excerpt":"Conventional wisdom holds that large-batch training is fundamentally incompatible with Reinforcement Learning (RL) - beyond a modest threshold, increasing batch sizes typically yields diminishing returns or performance degradation due to the inherent non-stationarity of the data distribution. We challenge this view by observing that non-stationarity is not a fixed property of RL, but evolves throughout training: early stages exhibit rapid behavioral shifts that demand small batches for plasticity, whereas late stages approach a quasi-stationary regime where large batches enable precise converg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21557","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/2605.21557/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":"2026-05-22T00:02:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9RImdfQSEml31osNvfT/Q3+CTqKYzHrIl7LxxEjf+i7p8PrSEXkv5WI02+ReHLz2Ort6fTdj95x7tZp/Y86wBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T15:23:50.805087Z"},"content_sha256":"6f88e1570f8b86a35b3f5442ac925a58ce0a1875aa239b840da9800c1918cc78","schema_version":"1.0","event_id":"sha256:6f88e1570f8b86a35b3f5442ac925a58ce0a1875aa239b840da9800c1918cc78"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CU2IPJNYP4JID5VBP3QMLAMOZM/bundle.json","state_url":"https://pith.science/pith/CU2IPJNYP4JID5VBP3QMLAMOZM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CU2IPJNYP4JID5VBP3QMLAMOZM/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":"2026-05-22T15:23:50Z","links":{"resolver":"https://pith.science/pith/CU2IPJNYP4JID5VBP3QMLAMOZM","bundle":"https://pith.science/pith/CU2IPJNYP4JID5VBP3QMLAMOZM/bundle.json","state":"https://pith.science/pith/CU2IPJNYP4JID5VBP3QMLAMOZM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CU2IPJNYP4JID5VBP3QMLAMOZM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CU2IPJNYP4JID5VBP3QMLAMOZM","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":"91d4d2d5322547fb031f3df2e193c64c01cf0d598dbb2929ec7f2a57a3aba1e7","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-20T13:46:22Z","title_canon_sha256":"539be7504c6a3b01f872918005e53a0f847c006e739455e7870b362a8c6ad34e"},"schema_version":"1.0","source":{"id":"2605.21557","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21557","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21557v1","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21557","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"pith_short_12","alias_value":"CU2IPJNYP4JI","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"pith_short_16","alias_value":"CU2IPJNYP4JID5VB","created_at":"2026-05-22T00:02:28Z"},{"alias_kind":"pith_short_8","alias_value":"CU2IPJNY","created_at":"2026-05-22T00:02:28Z"}],"graph_snapshots":[{"event_id":"sha256:6f88e1570f8b86a35b3f5442ac925a58ce0a1875aa239b840da9800c1918cc78","target":"graph","created_at":"2026-05-22T00:02:28Z","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/2605.21557/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Conventional wisdom holds that large-batch training is fundamentally incompatible with Reinforcement Learning (RL) - beyond a modest threshold, increasing batch sizes typically yields diminishing returns or performance degradation due to the inherent non-stationarity of the data distribution. We challenge this view by observing that non-stationarity is not a fixed property of RL, but evolves throughout training: early stages exhibit rapid behavioral shifts that demand small batches for plasticity, whereas late stages approach a quasi-stationary regime where large batches enable precise converg","authors_text":"Jongchan Park","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-20T13:46:22Z","title":"Scalable On-Policy Reinforcement Learning via Adaptive Batch Scaling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21557","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":"sha256:0ca42d388553016acf1aaeda422b1c0afb64a7e0a81d600e6edf010f845dd2ca","target":"record","created_at":"2026-05-22T00:02:28Z","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":"91d4d2d5322547fb031f3df2e193c64c01cf0d598dbb2929ec7f2a57a3aba1e7","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-20T13:46:22Z","title_canon_sha256":"539be7504c6a3b01f872918005e53a0f847c006e739455e7870b362a8c6ad34e"},"schema_version":"1.0","source":{"id":"2605.21557","kind":"arxiv","version":1}},"canonical_sha256":"153487a5b87f1281f6a17ee0c5818ecb3f0a656cb52a2413c496bb47601acab7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"153487a5b87f1281f6a17ee0c5818ecb3f0a656cb52a2413c496bb47601acab7","first_computed_at":"2026-05-22T00:02:28.410043Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T00:02:28.410043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/QdExPrIAmkKeo2z64N59hKr5CcfxTJY7UbWTVjEAXGRgpuLl9akk6n0BhrANvuSEXkeOXoj3LrMdK5rwns9Aw==","signature_status":"signed_v1","signed_at":"2026-05-22T00:02:28.410596Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.21557","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0ca42d388553016acf1aaeda422b1c0afb64a7e0a81d600e6edf010f845dd2ca","sha256:6f88e1570f8b86a35b3f5442ac925a58ce0a1875aa239b840da9800c1918cc78"],"state_sha256":"98b5146f1f53b0475be57e4c126b360eea0434dd861dd6f499f8c5974714dc1c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HxKY4fhcBiC8qlkUyMVuQ3IiH3V9pkyi5OSxumNvDJWPjSTG+UECRjAfClzTEr2UcS9Y7uvjuUDk3Ayg0AiYAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-22T15:23:50.808035Z","bundle_sha256":"3c89f193396b2842a1f791972d66aae543d8fdd82aae42098a28694ef987b116"}}