{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:LHJK6CGE45Z3KCYKTKXPJBNKGT","short_pith_number":"pith:LHJK6CGE","schema_version":"1.0","canonical_sha256":"59d2af08c4e773b50b0a9aaef485aa34ec25759ae380d53c4c091a8805e5d122","source":{"kind":"arxiv","id":"2212.05698","version":1},"attestation_state":"computed","paper":{"title":"MoDem: Accelerating Visual Model-Based Reinforcement Learning with Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Aravind Rajeswaran, Hao Su, Nicklas Hansen, Vikash Kumar, Xiaolong Wang, Yixin Lin","submitted_at":"2022-12-12T04:28:50Z","abstract_excerpt":"Poor sample efficiency continues to be the primary challenge for deployment of deep Reinforcement Learning (RL) algorithms for real-world applications, and in particular for visuo-motor control. Model-based RL has the potential to be highly sample efficient by concurrently learning a world model and using synthetic rollouts for planning and policy improvement. However, in practice, sample-efficient learning with model-based RL is bottlenecked by the exploration challenge. In this work, we find that leveraging just a handful of demonstrations can dramatically improve the sample-efficiency of mo"},"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":"2212.05698","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-12T04:28:50Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"c9e8b81d6dc049239f5a8d8c0b726e5f0be6a01f5abb11db0a14cbfd8c69158a","abstract_canon_sha256":"57807b1ebfde20342744861dbf760cdd286ed12e60e4b1cf395fe109c077c5b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:24:18.632059Z","signature_b64":"SFhLvuTPGh2W3dSwhtpwWWTvIVBFXWxmE+96IMe5H5/Vi6SvGboIHe19NFDY5Pv3GZP25UlAjc1eAO3VRZFzBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59d2af08c4e773b50b0a9aaef485aa34ec25759ae380d53c4c091a8805e5d122","last_reissued_at":"2026-07-05T05:24:18.631589Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:24:18.631589Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MoDem: Accelerating Visual Model-Based Reinforcement Learning with Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Aravind Rajeswaran, Hao Su, Nicklas Hansen, Vikash Kumar, Xiaolong Wang, Yixin Lin","submitted_at":"2022-12-12T04:28:50Z","abstract_excerpt":"Poor sample efficiency continues to be the primary challenge for deployment of deep Reinforcement Learning (RL) algorithms for real-world applications, and in particular for visuo-motor control. Model-based RL has the potential to be highly sample efficient by concurrently learning a world model and using synthetic rollouts for planning and policy improvement. However, in practice, sample-efficient learning with model-based RL is bottlenecked by the exploration challenge. In this work, we find that leveraging just a handful of demonstrations can dramatically improve the sample-efficiency of mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.05698","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/2212.05698/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":"2212.05698","created_at":"2026-07-05T05:24:18.631647+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.05698v1","created_at":"2026-07-05T05:24:18.631647+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.05698","created_at":"2026-07-05T05:24:18.631647+00:00"},{"alias_kind":"pith_short_12","alias_value":"LHJK6CGE45Z3","created_at":"2026-07-05T05:24:18.631647+00:00"},{"alias_kind":"pith_short_16","alias_value":"LHJK6CGE45Z3KCYK","created_at":"2026-07-05T05:24:18.631647+00:00"},{"alias_kind":"pith_short_8","alias_value":"LHJK6CGE","created_at":"2026-07-05T05:24:18.631647+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18888","citing_title":"Generative-Model Predictive Planning for Navigation in Partially Observable Environments","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2507.07986","citing_title":"EXPO: Stable Reinforcement Learning with Expressive Policies","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11075","citing_title":"RISE: Self-Improving Robot Policy with Compositional World Model","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT","json":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT.json","graph_json":"https://pith.science/api/pith-number/LHJK6CGE45Z3KCYKTKXPJBNKGT/graph.json","events_json":"https://pith.science/api/pith-number/LHJK6CGE45Z3KCYKTKXPJBNKGT/events.json","paper":"https://pith.science/paper/LHJK6CGE"},"agent_actions":{"view_html":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT","download_json":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT.json","view_paper":"https://pith.science/paper/LHJK6CGE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.05698&json=true","fetch_graph":"https://pith.science/api/pith-number/LHJK6CGE45Z3KCYKTKXPJBNKGT/graph.json","fetch_events":"https://pith.science/api/pith-number/LHJK6CGE45Z3KCYKTKXPJBNKGT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT/action/storage_attestation","attest_author":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT/action/author_attestation","sign_citation":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT/action/citation_signature","submit_replication":"https://pith.science/pith/LHJK6CGE45Z3KCYKTKXPJBNKGT/action/replication_record"}},"created_at":"2026-07-05T05:24:18.631647+00:00","updated_at":"2026-07-05T05:24:18.631647+00:00"}