{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:CZFYMMBGPHE76TSZLYQ4KCNGMP","short_pith_number":"pith:CZFYMMBG","canonical_record":{"source":{"id":"2510.01460","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-01T20:58:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5358f4eaa605e52c7d4a6f166aa4c6c6e9054b289bb2ef688f08b98ca01cacc5","abstract_canon_sha256":"8c52c0124b1aa2dd7455c67090700914c65295d8d8f44d04bd430435c6a84cc3"},"schema_version":"1.0"},"canonical_sha256":"164b86302679c9ff4e595e21c509a663d06519848a7f053d9766ba9ab0e2c52c","source":{"kind":"arxiv","id":"2510.01460","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.01460","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"arxiv_version","alias_value":"2510.01460v4","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.01460","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"pith_short_12","alias_value":"CZFYMMBGPHE7","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"pith_short_16","alias_value":"CZFYMMBGPHE76TSZ","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"pith_short_8","alias_value":"CZFYMMBG","created_at":"2026-07-07T02:17:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:CZFYMMBGPHE76TSZLYQ4KCNGMP","target":"record","payload":{"canonical_record":{"source":{"id":"2510.01460","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-01T20:58:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5358f4eaa605e52c7d4a6f166aa4c6c6e9054b289bb2ef688f08b98ca01cacc5","abstract_canon_sha256":"8c52c0124b1aa2dd7455c67090700914c65295d8d8f44d04bd430435c6a84cc3"},"schema_version":"1.0"},"canonical_sha256":"164b86302679c9ff4e595e21c509a663d06519848a7f053d9766ba9ab0e2c52c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:17:10.836242Z","signature_b64":"YnS00/vM5NqFB80XOEM/7cmOy/wK/u7qNoGy2BsVdycD5EUNI7wD4xTm99WI16upXjHWK1wfBtgRDUxNw3siCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"164b86302679c9ff4e595e21c509a663d06519848a7f053d9766ba9ab0e2c52c","last_reissued_at":"2026-07-07T02:17:10.835346Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:17:10.835346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2510.01460","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-07T02:17:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iZ4tcKpmxBU3VOIGMBBdM2DJQrUUWyCyLXDcZzIoHRM3A9oC/lxYK0MKUwumqcIUN0DCZ+DDoT/46unxuZakDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T01:03:10.341512Z"},"content_sha256":"7577a6f4e162220624a162e8a47ba41c8634a005389557f88dd9d9f94b35a0ec","schema_version":"1.0","event_id":"sha256:7577a6f4e162220624a162e8a47ba41c8634a005389557f88dd9d9f94b35a0ec"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:CZFYMMBGPHE76TSZLYQ4KCNGMP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Three Regimes of Offline-to-Online Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Lu Li, Pierre-Luc Bacon, Tianwei Ni, Yihao Sun","submitted_at":"2025-10-01T20:58:14Z","abstract_excerpt":"Offline-to-online reinforcement learning (RL) has emerged as a practical paradigm that leverages offline datasets for pretraining and online interactions for fine-tuning. However, its empirical behavior is highly inconsistent: design choices of online fine-tuning that work well in one setting can fail completely in another. Guided by the stability--plasticity principle, we propose a framework that can explain this inconsistency: We argue that efficient fine-tuning must preserve the utility of the stronger offline prior, whether that is the pretrained policy or the offline dataset, while mainta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.01460","kind":"arxiv","version":4},"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/2510.01460/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":"2026-07-07T02:17:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"osTKd8NJdrfW6i5g/Wnz0tnMmIuv6PZzNzi0PURMou5dVFxPcy7UAfAXQyrEne+vftCacOT7BZThhFVqsBBzDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T01:03:10.342086Z"},"content_sha256":"7239dceef1d7186cbee699962d1090c7108668110c037b517d379eb56d031a3f","schema_version":"1.0","event_id":"sha256:7239dceef1d7186cbee699962d1090c7108668110c037b517d379eb56d031a3f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP/bundle.json","state_url":"https://pith.science/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP/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":"2026-08-06T01:03:10Z","links":{"resolver":"https://pith.science/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP","bundle":"https://pith.science/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP/bundle.json","state":"https://pith.science/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CZFYMMBGPHE76TSZLYQ4KCNGMP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:CZFYMMBGPHE76TSZLYQ4KCNGMP","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":"8c52c0124b1aa2dd7455c67090700914c65295d8d8f44d04bd430435c6a84cc3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-01T20:58:14Z","title_canon_sha256":"5358f4eaa605e52c7d4a6f166aa4c6c6e9054b289bb2ef688f08b98ca01cacc5"},"schema_version":"1.0","source":{"id":"2510.01460","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.01460","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"arxiv_version","alias_value":"2510.01460v4","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.01460","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"pith_short_12","alias_value":"CZFYMMBGPHE7","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"pith_short_16","alias_value":"CZFYMMBGPHE76TSZ","created_at":"2026-07-07T02:17:10Z"},{"alias_kind":"pith_short_8","alias_value":"CZFYMMBG","created_at":"2026-07-07T02:17:10Z"}],"graph_snapshots":[{"event_id":"sha256:7239dceef1d7186cbee699962d1090c7108668110c037b517d379eb56d031a3f","target":"graph","created_at":"2026-07-07T02:17:10Z","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/2510.01460/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Offline-to-online reinforcement learning (RL) has emerged as a practical paradigm that leverages offline datasets for pretraining and online interactions for fine-tuning. However, its empirical behavior is highly inconsistent: design choices of online fine-tuning that work well in one setting can fail completely in another. Guided by the stability--plasticity principle, we propose a framework that can explain this inconsistency: We argue that efficient fine-tuning must preserve the utility of the stronger offline prior, whether that is the pretrained policy or the offline dataset, while mainta","authors_text":"Lu Li, Pierre-Luc Bacon, Tianwei Ni, Yihao Sun","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-01T20:58:14Z","title":"The Three Regimes of Offline-to-Online Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.01460","kind":"arxiv","version":4},"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":"sha256:7577a6f4e162220624a162e8a47ba41c8634a005389557f88dd9d9f94b35a0ec","target":"record","created_at":"2026-07-07T02:17:10Z","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":"8c52c0124b1aa2dd7455c67090700914c65295d8d8f44d04bd430435c6a84cc3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-01T20:58:14Z","title_canon_sha256":"5358f4eaa605e52c7d4a6f166aa4c6c6e9054b289bb2ef688f08b98ca01cacc5"},"schema_version":"1.0","source":{"id":"2510.01460","kind":"arxiv","version":4}},"canonical_sha256":"164b86302679c9ff4e595e21c509a663d06519848a7f053d9766ba9ab0e2c52c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"164b86302679c9ff4e595e21c509a663d06519848a7f053d9766ba9ab0e2c52c","first_computed_at":"2026-07-07T02:17:10.835346Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-07T02:17:10.835346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YnS00/vM5NqFB80XOEM/7cmOy/wK/u7qNoGy2BsVdycD5EUNI7wD4xTm99WI16upXjHWK1wfBtgRDUxNw3siCA==","signature_status":"signed_v1","signed_at":"2026-07-07T02:17:10.836242Z","signed_message":"canonical_sha256_bytes"},"source_id":"2510.01460","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7577a6f4e162220624a162e8a47ba41c8634a005389557f88dd9d9f94b35a0ec","sha256:7239dceef1d7186cbee699962d1090c7108668110c037b517d379eb56d031a3f"],"state_sha256":"7949b3d7493cfe894eef50b4706032db77e7a490333b29e78ad9088166d56491"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sAHdPUsRBuN2EIFLFBODLBM3F1IxoFzPE9WEDpQsSrjgZ8xPrutjm6b/ejohUWkOQKKxCQGpS2uKhZCeFeZgDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T01:03:10.348285Z","bundle_sha256":"ef0693d826a40fb44be175770185993089096a5f5e44447624faa42233e07d70"}}