{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OY6IPXJIIYRZV4ZY4BRL6P4DH5","short_pith_number":"pith:OY6IPXJI","schema_version":"1.0","canonical_sha256":"763c87dd2846239af338e062bf3f833f782e31e3a05b7df20eedb2e0a030aaae","source":{"kind":"arxiv","id":"2505.08361","version":1},"attestation_state":"computed","paper":{"title":"Modeling Unseen Environments with Language-guided Composable Causal Components in Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Biwei Huang, Xinyue Wang","submitted_at":"2025-05-13T09:08:28Z","abstract_excerpt":"Generalization in reinforcement learning (RL) remains a significant challenge, especially when agents encounter novel environments with unseen dynamics. Drawing inspiration from human compositional reasoning -- where known components are reconfigured to handle new situations -- we introduce World Modeling with Compositional Causal Components (WM3C). This novel framework enhances RL generalization by learning and leveraging compositional causal components. Unlike previous approaches focusing on invariant representation learning or meta-learning, WM3C identifies and utilizes causal dynamics amon"},"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":"2505.08361","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-13T09:08:28Z","cross_cats_sorted":[],"title_canon_sha256":"965f9b1695399fa3f8a970886e7f95908bd4509558f958be317b2a7fbdc41819","abstract_canon_sha256":"34e0e51d31ab466608a3b66bc5882f126e4d503478971ac8449d0f3e63447662"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:29.498860Z","signature_b64":"C8STy804MExMQIE1P08tLPPElZc5XsLR+bm1ILkU+QpyZzN3XLmOvHLEg/kJWsafKQ3WSrkGPTfSjf5elvfHDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"763c87dd2846239af338e062bf3f833f782e31e3a05b7df20eedb2e0a030aaae","last_reissued_at":"2026-07-05T11:02:29.498410Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:29.498410Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modeling Unseen Environments with Language-guided Composable Causal Components in Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Biwei Huang, Xinyue Wang","submitted_at":"2025-05-13T09:08:28Z","abstract_excerpt":"Generalization in reinforcement learning (RL) remains a significant challenge, especially when agents encounter novel environments with unseen dynamics. Drawing inspiration from human compositional reasoning -- where known components are reconfigured to handle new situations -- we introduce World Modeling with Compositional Causal Components (WM3C). This novel framework enhances RL generalization by learning and leveraging compositional causal components. Unlike previous approaches focusing on invariant representation learning or meta-learning, WM3C identifies and utilizes causal dynamics amon"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08361","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/2505.08361/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":"2505.08361","created_at":"2026-07-05T11:02:29.498468+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08361v1","created_at":"2026-07-05T11:02:29.498468+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08361","created_at":"2026-07-05T11:02:29.498468+00:00"},{"alias_kind":"pith_short_12","alias_value":"OY6IPXJIIYRZ","created_at":"2026-07-05T11:02:29.498468+00:00"},{"alias_kind":"pith_short_16","alias_value":"OY6IPXJIIYRZV4ZY","created_at":"2026-07-05T11:02:29.498468+00:00"},{"alias_kind":"pith_short_8","alias_value":"OY6IPXJI","created_at":"2026-07-05T11:02:29.498468+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16054","citing_title":"Ada-Diffuser: Latent-Aware Adaptive Diffusion for Decision-Making","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10118","citing_title":"Plan in Sandbox, Navigate in Open Worlds: Learning Physics-Grounded Abstracted Experience for Embodied Navigation","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02681","citing_title":"The Design and Composition of Structural Causal Decision Processes","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5","json":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5.json","graph_json":"https://pith.science/api/pith-number/OY6IPXJIIYRZV4ZY4BRL6P4DH5/graph.json","events_json":"https://pith.science/api/pith-number/OY6IPXJIIYRZV4ZY4BRL6P4DH5/events.json","paper":"https://pith.science/paper/OY6IPXJI"},"agent_actions":{"view_html":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5","download_json":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5.json","view_paper":"https://pith.science/paper/OY6IPXJI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08361&json=true","fetch_graph":"https://pith.science/api/pith-number/OY6IPXJIIYRZV4ZY4BRL6P4DH5/graph.json","fetch_events":"https://pith.science/api/pith-number/OY6IPXJIIYRZV4ZY4BRL6P4DH5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5/action/storage_attestation","attest_author":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5/action/author_attestation","sign_citation":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5/action/citation_signature","submit_replication":"https://pith.science/pith/OY6IPXJIIYRZV4ZY4BRL6P4DH5/action/replication_record"}},"created_at":"2026-07-05T11:02:29.498468+00:00","updated_at":"2026-07-05T11:02:29.498468+00:00"}