{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3BDUU32BFAZ77BT27T4RFR26AE","short_pith_number":"pith:3BDUU32B","schema_version":"1.0","canonical_sha256":"d8474a6f412833ff867afcf912c75e012127ad12dbb5d9f57a93f3b4263df78c","source":{"kind":"arxiv","id":"2410.05364","version":2},"attestation_state":"computed","paper":{"title":"Diffusion Model Predictive Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Antoine Dedieu, Guangyao Zhou, Joseph Ortiz, J. Swaroop Guntupalli, Kevin Murphy, Miguel L\\'azaro-Gredilla, Rajkumar Vasudeva Raju, Sivaramakrishnan Swaminathan, Wolfgang Lehrach","submitted_at":"2024-10-07T17:56:47Z","abstract_excerpt":"We propose Diffusion Model Predictive Control (D-MPC), a novel MPC approach that learns a multi-step action proposal and a multi-step dynamics model, both using diffusion models, and combines them for use in online MPC. On the popular D4RL benchmark, we show performance that is significantly better than existing model-based offline planning methods using MPC (e.g. MBOP) and competitive with state-of-the-art (SOTA) model-based and model-free reinforcement learning methods. We additionally illustrate D-MPC's ability to optimize novel reward functions at run time and adapt to novel dynamics, and "},"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":"2410.05364","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T17:56:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"77e3981f7ef4d71d2fa3ab02affefd0d4e377ebcb28205c1662d27a49078124a","abstract_canon_sha256":"77ae59c026930d2d2166d2f79b6b32d07851eed2ff1be912ee23b634f9f60829"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:01.583025Z","signature_b64":"Dnxk/Eg7FXqVytCp9sEvpdKeT9APFWFBWgz+fsXqauLs1NUzHr9PNn7TcEQrkHSH28aETryArMabc90kUhkgDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8474a6f412833ff867afcf912c75e012127ad12dbb5d9f57a93f3b4263df78c","last_reissued_at":"2026-07-05T11:08:01.582529Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:01.582529Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diffusion Model Predictive Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Antoine Dedieu, Guangyao Zhou, Joseph Ortiz, J. Swaroop Guntupalli, Kevin Murphy, Miguel L\\'azaro-Gredilla, Rajkumar Vasudeva Raju, Sivaramakrishnan Swaminathan, Wolfgang Lehrach","submitted_at":"2024-10-07T17:56:47Z","abstract_excerpt":"We propose Diffusion Model Predictive Control (D-MPC), a novel MPC approach that learns a multi-step action proposal and a multi-step dynamics model, both using diffusion models, and combines them for use in online MPC. On the popular D4RL benchmark, we show performance that is significantly better than existing model-based offline planning methods using MPC (e.g. MBOP) and competitive with state-of-the-art (SOTA) model-based and model-free reinforcement learning methods. We additionally illustrate D-MPC's ability to optimize novel reward functions at run time and adapt to novel dynamics, and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05364","kind":"arxiv","version":2},"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/2410.05364/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":"2410.05364","created_at":"2026-07-05T11:08:01.582590+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.05364v2","created_at":"2026-07-05T11:08:01.582590+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05364","created_at":"2026-07-05T11:08:01.582590+00:00"},{"alias_kind":"pith_short_12","alias_value":"3BDUU32BFAZ7","created_at":"2026-07-05T11:08:01.582590+00:00"},{"alias_kind":"pith_short_16","alias_value":"3BDUU32BFAZ77BT2","created_at":"2026-07-05T11:08:01.582590+00:00"},{"alias_kind":"pith_short_8","alias_value":"3BDUU32B","created_at":"2026-07-05T11:08:01.582590+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18888","citing_title":"Generative-Model Predictive Planning for Navigation in Partially Observable Environments","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10825","citing_title":"MODIP: Efficient Model-Based Optimization for Diffusion Policies","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2510.04280","citing_title":"A KL-regularization Framework for Learning to Plan with Adaptive Priors","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2512.24497","citing_title":"What Drives Success in Physical Planning with Joint-Embedding Predictive World Models?","ref_index":78,"is_internal_anchor":false},{"citing_arxiv_id":"2509.19538","citing_title":"DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2603.17834","citing_title":"Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27450","citing_title":"RAY-TOLD: Ray-Based Latent Dynamics for Dense Dynamic Obstacle Avoidance with TDMPC","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18343","citing_title":"DAG-STL: A Hierarchical Framework for Zero-Shot Trajectory Planning under Signal Temporal Logic Specifications","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07457","citing_title":"CMP: Robust Whole-Body Tracking for Loco-Manipulation via Competence Manifold Projection","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07084","citing_title":"Flow Motion Policy: Manipulator Motion Planning with Flow Matching Models","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06133","citing_title":"Learning-Guided Force-Feedback Model Predictive Control with Obstacle Avoidance for Robotic Deburring","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE","json":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE.json","graph_json":"https://pith.science/api/pith-number/3BDUU32BFAZ77BT27T4RFR26AE/graph.json","events_json":"https://pith.science/api/pith-number/3BDUU32BFAZ77BT27T4RFR26AE/events.json","paper":"https://pith.science/paper/3BDUU32B"},"agent_actions":{"view_html":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE","download_json":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE.json","view_paper":"https://pith.science/paper/3BDUU32B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.05364&json=true","fetch_graph":"https://pith.science/api/pith-number/3BDUU32BFAZ77BT27T4RFR26AE/graph.json","fetch_events":"https://pith.science/api/pith-number/3BDUU32BFAZ77BT27T4RFR26AE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE/action/storage_attestation","attest_author":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE/action/author_attestation","sign_citation":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE/action/citation_signature","submit_replication":"https://pith.science/pith/3BDUU32BFAZ77BT27T4RFR26AE/action/replication_record"}},"created_at":"2026-07-05T11:08:01.582590+00:00","updated_at":"2026-07-05T11:08:01.582590+00:00"}