{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DUU2F4QQDSFGQGTJ2OSJR55MBW","short_pith_number":"pith:DUU2F4QQ","schema_version":"1.0","canonical_sha256":"1d29a2f2101c8a681a69d3a498f7ac0d842f26b4855027d09dc776048dd26d11","source":{"kind":"arxiv","id":"2310.00344","version":3},"attestation_state":"computed","paper":{"title":"HarmonyDream: Task Harmonization Inside World Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chenjun Xiao, Dong Li, Haoyu Ma, Jialong Wu, Jianmin Wang, Jianye Hao, Mingsheng Long, Ningya Feng","submitted_at":"2023-09-30T11:38:13Z","abstract_excerpt":"Model-based reinforcement learning (MBRL) holds the promise of sample-efficient learning by utilizing a world model, which models how the environment works and typically encompasses components for two tasks: observation modeling and reward modeling. In this paper, through a dedicated empirical investigation, we gain a deeper understanding of the role each task plays in world models and uncover the overlooked potential of sample-efficient MBRL by mitigating the domination of either observation or reward modeling. Our key insight is that while prevalent approaches of explicit MBRL attempt to res"},"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":"2310.00344","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-30T11:38:13Z","cross_cats_sorted":[],"title_canon_sha256":"98cf6fca8c27fedef244c5df872c33e77db755ab11f193d4d8b41d205c1729b6","abstract_canon_sha256":"0f75026b3a6de1bb66d00484e67c3054c4df4af5cfd903a7cb10e324ada2c82b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:42.513773Z","signature_b64":"PPXeOQ/gQqH7K69dM6SAF5KeUuflrwiAR2wMDUefDrvvnKmvCsVflbHgFebLEY+emO/RcdwzcoklemT2KFPqDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d29a2f2101c8a681a69d3a498f7ac0d842f26b4855027d09dc776048dd26d11","last_reissued_at":"2026-07-05T08:27:42.513226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:42.513226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HarmonyDream: Task Harmonization Inside World Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chenjun Xiao, Dong Li, Haoyu Ma, Jialong Wu, Jianmin Wang, Jianye Hao, Mingsheng Long, Ningya Feng","submitted_at":"2023-09-30T11:38:13Z","abstract_excerpt":"Model-based reinforcement learning (MBRL) holds the promise of sample-efficient learning by utilizing a world model, which models how the environment works and typically encompasses components for two tasks: observation modeling and reward modeling. In this paper, through a dedicated empirical investigation, we gain a deeper understanding of the role each task plays in world models and uncover the overlooked potential of sample-efficient MBRL by mitigating the domination of either observation or reward modeling. Our key insight is that while prevalent approaches of explicit MBRL attempt to res"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00344","kind":"arxiv","version":3},"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/2310.00344/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":"2310.00344","created_at":"2026-07-05T08:27:42.513286+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.00344v3","created_at":"2026-07-05T08:27:42.513286+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00344","created_at":"2026-07-05T08:27:42.513286+00:00"},{"alias_kind":"pith_short_12","alias_value":"DUU2F4QQDSFG","created_at":"2026-07-05T08:27:42.513286+00:00"},{"alias_kind":"pith_short_16","alias_value":"DUU2F4QQDSFGQGTJ","created_at":"2026-07-05T08:27:42.513286+00:00"},{"alias_kind":"pith_short_8","alias_value":"DUU2F4QQ","created_at":"2026-07-05T08:27:42.513286+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12312","citing_title":"Transferable Delay-Aware Reinforcement Learning via Implicit Causal Graph Modeling","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW","json":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW.json","graph_json":"https://pith.science/api/pith-number/DUU2F4QQDSFGQGTJ2OSJR55MBW/graph.json","events_json":"https://pith.science/api/pith-number/DUU2F4QQDSFGQGTJ2OSJR55MBW/events.json","paper":"https://pith.science/paper/DUU2F4QQ"},"agent_actions":{"view_html":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW","download_json":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW.json","view_paper":"https://pith.science/paper/DUU2F4QQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.00344&json=true","fetch_graph":"https://pith.science/api/pith-number/DUU2F4QQDSFGQGTJ2OSJR55MBW/graph.json","fetch_events":"https://pith.science/api/pith-number/DUU2F4QQDSFGQGTJ2OSJR55MBW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW/action/storage_attestation","attest_author":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW/action/author_attestation","sign_citation":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW/action/citation_signature","submit_replication":"https://pith.science/pith/DUU2F4QQDSFGQGTJ2OSJR55MBW/action/replication_record"}},"created_at":"2026-07-05T08:27:42.513286+00:00","updated_at":"2026-07-05T08:27:42.513286+00:00"}