{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XE4DZ72WMHQWO2RL4XB2E2ZYSN","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":"0aa12db3538b1b684f3c6617c97b68a99d7954b94c3f0e293508d2cddd1ee5ff","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-04-18T17:58:03Z","title_canon_sha256":"857a47855d259cc296814b1c63587a86ee8d37905cfeb28ac34dabab669fa6c2"},"schema_version":"1.0","source":{"id":"2404.12377","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.12377","created_at":"2026-05-17T23:38:50Z"},{"alias_kind":"arxiv_version","alias_value":"2404.12377v1","created_at":"2026-05-17T23:38:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.12377","created_at":"2026-05-17T23:38:50Z"},{"alias_kind":"pith_short_12","alias_value":"XE4DZ72WMHQW","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"XE4DZ72WMHQWO2RL","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"XE4DZ72W","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:02f86d6b471ae03ab373da35908a96a87558e71a3cd0f5ce29866f9186500ac6","target":"graph","created_at":"2026-05-17T23:38:50Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our approach can successfully synthesize video plans on unseen goals in the RT-X, enables successful robot execution in simulation, and substantially outperforms monolithic baseline approaches to video generation."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That natural language instructions can be reliably parsed into lower-level primitives whose separate models compose into coherent, realistic videos without introducing artifacts or losing task-relevant details."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"RoboDreamer factorizes video generation using language primitives to achieve compositional generalization in robot world models, outperforming monolithic baselines on unseen goals in RT-X."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"RoboDreamer factorizes video generation using language primitives to create plans for unseen robot tasks."}],"snapshot_sha256":"98546809f5544792a6442e94ac62eae53b46ea6b6b1221ea558c9a5d60fcd642"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"c256a1870659fa9e3d01b953c60dd7a4b6a37d8529245a7b3f095707f4136afd"},"paper":{"abstract_excerpt":"Text-to-video models have demonstrated substantial potential in robotic decision-making, enabling the imagination of realistic plans of future actions as well as accurate environment simulation. However, one major issue in such models is generalization -- models are limited to synthesizing videos subject to language instructions similar to those seen at training time. This is heavily limiting in decision-making, where we seek a powerful world model to synthesize plans of unseen combinations of objects and actions in order to solve previously unseen tasks in new environments. To resolve this is","authors_text":"Chuang Gan, Dit-Yan Yeung, Jiaben Chen, Siyuan Zhou, Yandong Li, Yilun Du","cross_cats":[],"headline":"RoboDreamer factorizes video generation using language primitives to create plans for unseen robot tasks.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-04-18T17:58:03Z","title":"RoboDreamer: Learning Compositional World Models for Robot Imagination"},"references":{"count":65,"internal_anchors":12,"resolved_work":65,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":7,"title":"Unsupervised learning of compositional energy concepts","work_id":"585f2b86-ddff-414d-ada6-8a36e14fef70","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":8,"title":"B., Dieleman, S., Fergus, R., Sohl-Dickstein, J., Doucet, A., and Grathwohl, W","work_id":"1bf90112-a203-409f-8900-11651a71919b","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":12,"title":"G., Tapaswi, M., Laptev, I., and Schmid, C","work_id":"646c92f1-d546-4fbb-8968-71d5ee6f9718","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":15,"title":"Diffusion-based generation, optimization, and planning in 3d scenes","work_id":"6bff28de-131c-49cc-86fa-2b61c4212c98","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":17,"title":"R., and Davison, A","work_id":"ebc7a356-db9d-48bd-8ae8-be7274070335","year":2020}],"snapshot_sha256":"d9f370a51d0cca10de936e6bd758f85f4a968cc6ab1eb5a137c4add3f2f0e6f7"},"source":{"id":"2404.12377","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T20:43:49.453064Z","id":"cd704efb-c6fd-468c-992c-3d2d32701d49","model_set":{"reader":"grok-4.3"},"one_line_summary":"RoboDreamer factorizes video generation using language primitives to achieve compositional generalization in robot world models, outperforming monolithic baselines on unseen goals in RT-X.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"RoboDreamer factorizes video generation using language primitives to create plans for unseen robot tasks.","strongest_claim":"Our approach can successfully synthesize video plans on unseen goals in the RT-X, enables successful robot execution in simulation, and substantially outperforms monolithic baseline approaches to video generation.","weakest_assumption":"That natural language instructions can be reliably parsed into lower-level primitives whose separate models compose into coherent, realistic videos without introducing artifacts or losing task-relevant details."}},"verdict_id":"cd704efb-c6fd-468c-992c-3d2d32701d49"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:62d2801ba00f028b10774dc447dce23fad9fc7b384ad5b2bcc65cb67a875ba23","target":"record","created_at":"2026-05-17T23:38:50Z","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":"0aa12db3538b1b684f3c6617c97b68a99d7954b94c3f0e293508d2cddd1ee5ff","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-04-18T17:58:03Z","title_canon_sha256":"857a47855d259cc296814b1c63587a86ee8d37905cfeb28ac34dabab669fa6c2"},"schema_version":"1.0","source":{"id":"2404.12377","kind":"arxiv","version":1}},"canonical_sha256":"b9383cff5661e1676a2be5c3a26b38935c6a5cae241ad73df8473f74c3e799dc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b9383cff5661e1676a2be5c3a26b38935c6a5cae241ad73df8473f74c3e799dc","first_computed_at":"2026-05-17T23:38:50.211982Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:50.211982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XT8L6YcDMlDMVRuxhrl0lwFVfYAvO93+oGD+2UKeliIJC+4Dch/TBJ3zUsF6tnqTMg4qmsqrVgxKh6CUIUKCAw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:50.212577Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.12377","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:62d2801ba00f028b10774dc447dce23fad9fc7b384ad5b2bcc65cb67a875ba23","sha256:02f86d6b471ae03ab373da35908a96a87558e71a3cd0f5ce29866f9186500ac6"],"state_sha256":"7bd9739e44c3f6a845133b329f107f051226562efc613a3a0225d706dfbdd988"}