{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6FRWTIG4SF72REZ724FQ6S74DE","short_pith_number":"pith:6FRWTIG4","schema_version":"1.0","canonical_sha256":"f16369a0dc917fa8933fd70b0f4bfc1936b0cdeb6f3d493a8ef6cd12580bdaed","source":{"kind":"arxiv","id":"2206.14244","version":3},"attestation_state":"computed","paper":{"title":"Masked World Models for Visual Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Danijar Hafner, Fangchen Liu, Hao Liu, Kimin Lee, Pieter Abbeel, Stephen James, Younggyo Seo","submitted_at":"2022-06-28T18:42:27Z","abstract_excerpt":"Visual model-based reinforcement learning (RL) has the potential to enable sample-efficient robot learning from visual observations. Yet the current approaches typically train a single model end-to-end for learning both visual representations and dynamics, making it difficult to accurately model the interaction between robots and small objects. In this work, we introduce a visual model-based RL framework that decouples visual representation learning and dynamics learning. Specifically, we train an autoencoder with convolutional layers and vision transformers (ViT) to reconstruct pixels given m"},"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":"2206.14244","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2022-06-28T18:42:27Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"2b9a0cc76944b8b5c55f3e47445e264ab8a0d743c135e5de197b6337f17177fc","abstract_canon_sha256":"bf9697d0ad4a2af6b07182c332636429776346d57dfde18ba1f096d868deecd0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:35.664939Z","signature_b64":"dYYkJ2pADRpu5NLjzO+3oHP/Hh7wTX1GplST/YmViPIcAb5koK9Ov7sjcfWCw37NJ9EWezgL7nmhn5oeMZH/CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f16369a0dc917fa8933fd70b0f4bfc1936b0cdeb6f3d493a8ef6cd12580bdaed","last_reissued_at":"2026-07-05T06:14:35.664433Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:35.664433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Masked World Models for Visual Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Danijar Hafner, Fangchen Liu, Hao Liu, Kimin Lee, Pieter Abbeel, Stephen James, Younggyo Seo","submitted_at":"2022-06-28T18:42:27Z","abstract_excerpt":"Visual model-based reinforcement learning (RL) has the potential to enable sample-efficient robot learning from visual observations. Yet the current approaches typically train a single model end-to-end for learning both visual representations and dynamics, making it difficult to accurately model the interaction between robots and small objects. In this work, we introduce a visual model-based RL framework that decouples visual representation learning and dynamics learning. Specifically, we train an autoencoder with convolutional layers and vision transformers (ViT) to reconstruct pixels given m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.14244","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/2206.14244/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":"2206.14244","created_at":"2026-07-05T06:14:35.664488+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.14244v3","created_at":"2026-07-05T06:14:35.664488+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.14244","created_at":"2026-07-05T06:14:35.664488+00:00"},{"alias_kind":"pith_short_12","alias_value":"6FRWTIG4SF72","created_at":"2026-07-05T06:14:35.664488+00:00"},{"alias_kind":"pith_short_16","alias_value":"6FRWTIG4SF72REZ7","created_at":"2026-07-05T06:14:35.664488+00:00"},{"alias_kind":"pith_short_8","alias_value":"6FRWTIG4","created_at":"2026-07-05T06:14:35.664488+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.14427","citing_title":"Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":294,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE","json":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE.json","graph_json":"https://pith.science/api/pith-number/6FRWTIG4SF72REZ724FQ6S74DE/graph.json","events_json":"https://pith.science/api/pith-number/6FRWTIG4SF72REZ724FQ6S74DE/events.json","paper":"https://pith.science/paper/6FRWTIG4"},"agent_actions":{"view_html":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE","download_json":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE.json","view_paper":"https://pith.science/paper/6FRWTIG4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.14244&json=true","fetch_graph":"https://pith.science/api/pith-number/6FRWTIG4SF72REZ724FQ6S74DE/graph.json","fetch_events":"https://pith.science/api/pith-number/6FRWTIG4SF72REZ724FQ6S74DE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE/action/storage_attestation","attest_author":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE/action/author_attestation","sign_citation":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE/action/citation_signature","submit_replication":"https://pith.science/pith/6FRWTIG4SF72REZ724FQ6S74DE/action/replication_record"}},"created_at":"2026-07-05T06:14:35.664488+00:00","updated_at":"2026-07-05T06:14:35.664488+00:00"}