{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WL4X7VKMHDJPVJUBIX4Y4QEBNY","short_pith_number":"pith:WL4X7VKM","schema_version":"1.0","canonical_sha256":"b2f97fd54c38d2faa68145f98e40816e32420fc0d58fc613c984de39a49a55a7","source":{"kind":"arxiv","id":"2404.10662","version":2},"attestation_state":"computed","paper":{"title":"Continual Offline Reinforcement Learning via Diffusion-based Dual Generative Replay","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chunlin Chen, Jinmei Liu, Shilin Zhang, Wenbin Li, Xiangyu Yue, Zhi Wang","submitted_at":"2024-04-16T15:39:11Z","abstract_excerpt":"We study continual offline reinforcement learning, a practical paradigm that facilitates forward transfer and mitigates catastrophic forgetting to tackle sequential offline tasks. We propose a dual generative replay framework that retains previous knowledge by concurrent replay of generated pseudo-data. First, we decouple the continual learning policy into a diffusion-based generative behavior model and a multi-head action evaluation model, allowing the policy to inherit distributional expressivity for encompassing a progressive range of diverse behaviors. Second, we train a task-conditioned d"},"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":"2404.10662","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-16T15:39:11Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f96f779eab56d748d13fd1b0ef9cae7eb4b8106fcef7fa847f3e82a7649c2c8f","abstract_canon_sha256":"91134db7a5517bbb99c43801932015bcd837062e4573b36e116fab646864595f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:09:30.574928Z","signature_b64":"rHpZQhAOw8Slu3UZYdNfVcKTxWs2inzXmTCHLTmjPPfx6ptoNevFvtxgHzFKw9bd4RsMCFgskWF71wLTgqPHDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2f97fd54c38d2faa68145f98e40816e32420fc0d58fc613c984de39a49a55a7","last_reissued_at":"2026-07-05T08:09:30.574494Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:09:30.574494Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Continual Offline Reinforcement Learning via Diffusion-based Dual Generative Replay","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chunlin Chen, Jinmei Liu, Shilin Zhang, Wenbin Li, Xiangyu Yue, Zhi Wang","submitted_at":"2024-04-16T15:39:11Z","abstract_excerpt":"We study continual offline reinforcement learning, a practical paradigm that facilitates forward transfer and mitigates catastrophic forgetting to tackle sequential offline tasks. We propose a dual generative replay framework that retains previous knowledge by concurrent replay of generated pseudo-data. First, we decouple the continual learning policy into a diffusion-based generative behavior model and a multi-head action evaluation model, allowing the policy to inherit distributional expressivity for encompassing a progressive range of diverse behaviors. Second, we train a task-conditioned d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.10662","kind":"arxiv","version":2},"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/2404.10662/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":"2404.10662","created_at":"2026-07-05T08:09:30.574553+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.10662v2","created_at":"2026-07-05T08:09:30.574553+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.10662","created_at":"2026-07-05T08:09:30.574553+00:00"},{"alias_kind":"pith_short_12","alias_value":"WL4X7VKMHDJP","created_at":"2026-07-05T08:09:30.574553+00:00"},{"alias_kind":"pith_short_16","alias_value":"WL4X7VKMHDJPVJUB","created_at":"2026-07-05T08:09:30.574553+00:00"},{"alias_kind":"pith_short_8","alias_value":"WL4X7VKM","created_at":"2026-07-05T08:09:30.574553+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.05932","citing_title":"Skill Expansion and Composition in Parameter Space","ref_index":2021,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY","json":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY.json","graph_json":"https://pith.science/api/pith-number/WL4X7VKMHDJPVJUBIX4Y4QEBNY/graph.json","events_json":"https://pith.science/api/pith-number/WL4X7VKMHDJPVJUBIX4Y4QEBNY/events.json","paper":"https://pith.science/paper/WL4X7VKM"},"agent_actions":{"view_html":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY","download_json":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY.json","view_paper":"https://pith.science/paper/WL4X7VKM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.10662&json=true","fetch_graph":"https://pith.science/api/pith-number/WL4X7VKMHDJPVJUBIX4Y4QEBNY/graph.json","fetch_events":"https://pith.science/api/pith-number/WL4X7VKMHDJPVJUBIX4Y4QEBNY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY/action/storage_attestation","attest_author":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY/action/author_attestation","sign_citation":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY/action/citation_signature","submit_replication":"https://pith.science/pith/WL4X7VKMHDJPVJUBIX4Y4QEBNY/action/replication_record"}},"created_at":"2026-07-05T08:09:30.574553+00:00","updated_at":"2026-07-05T08:09:30.574553+00:00"}