{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UZUYT4FVUE45LOR4CEYGVO3XUD","short_pith_number":"pith:UZUYT4FV","schema_version":"1.0","canonical_sha256":"a66989f0b5a139d5ba3c11306abb77a0df80a7f89b45496496e80f2301890c71","source":{"kind":"arxiv","id":"2502.20476","version":2},"attestation_state":"computed","paper":{"title":"Unifying Model Predictive Path Integral Control, Reinforcement Learning, and Diffusion Models for Optimal Control and Planning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mo Chen, Yankai Li","submitted_at":"2025-02-27T19:26:36Z","abstract_excerpt":"Model Predictive Path Integral (MPPI) control, Reinforcement Learning (RL), and Diffusion Models have each demonstrated strong performance in trajectory optimization, decision-making, and motion planning. However, these approaches have traditionally been treated as distinct methodologies with separate optimization frameworks. In this work, we establish a unified perspective that connects MPPI, RL, and Diffusion Models through gradient-based optimization on the Gibbs measure. We first show that MPPI can be interpreted as performing gradient ascent on a smoothed energy function. We then demonstr"},"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":"2502.20476","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T19:26:36Z","cross_cats_sorted":[],"title_canon_sha256":"8f4c5379974c41c6a259447c9bd840ba47193c7255dcc060222915e2b6bdfc85","abstract_canon_sha256":"26c6573ef3f1def1ce2d0ce8fba45c0e464e3ac9d943e35e6d9bbf2922d0c8f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:24:03.092683Z","signature_b64":"ghbl/tNMaj5xky441pcamauIHX+Xx03XL1+SocwIhLKHkP7+STRZ8P71JH3j3y5FkAT8N4p8ap5OCgNBF/NcBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a66989f0b5a139d5ba3c11306abb77a0df80a7f89b45496496e80f2301890c71","last_reissued_at":"2026-07-05T10:24:03.092052Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:24:03.092052Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unifying Model Predictive Path Integral Control, Reinforcement Learning, and Diffusion Models for Optimal Control and Planning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mo Chen, Yankai Li","submitted_at":"2025-02-27T19:26:36Z","abstract_excerpt":"Model Predictive Path Integral (MPPI) control, Reinforcement Learning (RL), and Diffusion Models have each demonstrated strong performance in trajectory optimization, decision-making, and motion planning. However, these approaches have traditionally been treated as distinct methodologies with separate optimization frameworks. In this work, we establish a unified perspective that connects MPPI, RL, and Diffusion Models through gradient-based optimization on the Gibbs measure. We first show that MPPI can be interpreted as performing gradient ascent on a smoothed energy function. We then demonstr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20476","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/2502.20476/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":"2502.20476","created_at":"2026-07-05T10:24:03.092137+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20476v2","created_at":"2026-07-05T10:24:03.092137+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20476","created_at":"2026-07-05T10:24:03.092137+00:00"},{"alias_kind":"pith_short_12","alias_value":"UZUYT4FVUE45","created_at":"2026-07-05T10:24:03.092137+00:00"},{"alias_kind":"pith_short_16","alias_value":"UZUYT4FVUE45LOR4","created_at":"2026-07-05T10:24:03.092137+00:00"},{"alias_kind":"pith_short_8","alias_value":"UZUYT4FV","created_at":"2026-07-05T10:24:03.092137+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2512.17091","citing_title":"Learning to Plan, Planning to Learn: Adaptive Hierarchical RL-MPC for Sample-Efficient Decision Making","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD","json":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD.json","graph_json":"https://pith.science/api/pith-number/UZUYT4FVUE45LOR4CEYGVO3XUD/graph.json","events_json":"https://pith.science/api/pith-number/UZUYT4FVUE45LOR4CEYGVO3XUD/events.json","paper":"https://pith.science/paper/UZUYT4FV"},"agent_actions":{"view_html":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD","download_json":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD.json","view_paper":"https://pith.science/paper/UZUYT4FV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20476&json=true","fetch_graph":"https://pith.science/api/pith-number/UZUYT4FVUE45LOR4CEYGVO3XUD/graph.json","fetch_events":"https://pith.science/api/pith-number/UZUYT4FVUE45LOR4CEYGVO3XUD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD/action/storage_attestation","attest_author":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD/action/author_attestation","sign_citation":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD/action/citation_signature","submit_replication":"https://pith.science/pith/UZUYT4FVUE45LOR4CEYGVO3XUD/action/replication_record"}},"created_at":"2026-07-05T10:24:03.092137+00:00","updated_at":"2026-07-05T10:24:03.092137+00:00"}