{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IYYGEQYIKJZ547255DSSWREN7Z","short_pith_number":"pith:IYYGEQYI","schema_version":"1.0","canonical_sha256":"46306243085273de7f5de8e52b448dfe7dcaee771d01773e0800d44bd945530b","source":{"kind":"arxiv","id":"2605.04568","version":2},"attestation_state":"computed","paper":{"title":"Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance.","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Jonathan Spieler, Sven Behnke","submitted_at":"2026-05-06T07:13:11Z","abstract_excerpt":"State-of-the-art model-based Reinforcement Learning (RL) approaches either use gradient-free, population-based methods for planning, learned policy networks, or a combination of policy networks and planning. Hybrid approaches that combine Model Predictive Control (MPC) with a learned model and a policy prior to leverage the advantages of both paradigms have shown promising results. However, these approaches typically rely on gradient-free optimization methods, which can be computationally expensive for high-dimensional control tasks. While gradient-based methods are a promising alternative, re"},"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":"2605.04568","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-06T07:13:11Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"10edec659906056595f8435feb3dabf1501bab40dc053cbdf663f40b8cd86fbc","abstract_canon_sha256":"b601aed29fbd7d25fadca9352d056cdb7c42af1fe561f93e9ca9cf19c2980a09"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:21.620637Z","signature_b64":"wegpn6Frnal9Jau6WP8K9EPBgwiC1A+THdiFF05svJXVJLUztfPyy2SgQq2CWf0RFUaS6/ogiphU+qkbpfJKAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46306243085273de7f5de8e52b448dfe7dcaee771d01773e0800d44bd945530b","last_reissued_at":"2026-05-25T02:01:21.619872Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:21.619872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance.","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Jonathan Spieler, Sven Behnke","submitted_at":"2026-05-06T07:13:11Z","abstract_excerpt":"State-of-the-art model-based Reinforcement Learning (RL) approaches either use gradient-free, population-based methods for planning, learned policy networks, or a combination of policy networks and planning. Hybrid approaches that combine Model Predictive Control (MPC) with a learned model and a policy prior to leverage the advantages of both paradigms have shown promising results. However, these approaches typically rely on gradient-free optimization methods, which can be computationally expensive for high-dimensional control tasks. While gradient-based methods are a promising alternative, re"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our results on 24 continuous control tasks show that Dream-MPC can significantly improve the performance of the underlying policy and can outperform gradient-free MPC and state-of-the-art baselines.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The learned world model is sufficiently accurate and differentiable that gradient ascent on rolled-out trajectories reliably improves performance without being misled by model errors or local optima.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Dream-MPC boosts underlying policies on 24 continuous control tasks by optimizing policy-generated trajectories with gradient ascent, uncertainty regularization, and temporal amortization inside a latent world model.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"12fdb68e98465d566cdab351d22705ffba0aa13bc225bbcf49e8ca24164555df"},"source":{"id":"2605.04568","kind":"arxiv","version":2},"verdict":{"id":"3e10bde3-1524-4c77-a33e-565509147905","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T16:27:37.476846Z","strongest_claim":"Our results on 24 continuous control tasks show that Dream-MPC can significantly improve the performance of the underlying policy and can outperform gradient-free MPC and state-of-the-art baselines.","one_line_summary":"Dream-MPC boosts underlying policies on 24 continuous control tasks by optimizing policy-generated trajectories with gradient ascent, uncertainty regularization, and temporal amortization inside a latent world model.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The learned world model is sufficiently accurate and differentiable that gradient ascent on rolled-out trajectories reliably improves performance without being misled by model errors or local optima.","pith_extraction_headline":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.04568/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T11:39:47.694956Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T22:31:20.097591Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T14:20:08.650861Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"7a5f27b019c08312cefe236fc31addcf375473790e0ef05a1917df7b1cec37de"},"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":"2605.04568","created_at":"2026-05-25T02:01:21.619987+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.04568v2","created_at":"2026-05-25T02:01:21.619987+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04568","created_at":"2026-05-25T02:01:21.619987+00:00"},{"alias_kind":"pith_short_12","alias_value":"IYYGEQYIKJZ5","created_at":"2026-05-25T02:01:21.619987+00:00"},{"alias_kind":"pith_short_16","alias_value":"IYYGEQYIKJZ54725","created_at":"2026-05-25T02:01:21.619987+00:00"},{"alias_kind":"pith_short_8","alias_value":"IYYGEQYI","created_at":"2026-05-25T02:01:21.619987+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z","json":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z.json","graph_json":"https://pith.science/api/pith-number/IYYGEQYIKJZ547255DSSWREN7Z/graph.json","events_json":"https://pith.science/api/pith-number/IYYGEQYIKJZ547255DSSWREN7Z/events.json","paper":"https://pith.science/paper/IYYGEQYI"},"agent_actions":{"view_html":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z","download_json":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z.json","view_paper":"https://pith.science/paper/IYYGEQYI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.04568&json=true","fetch_graph":"https://pith.science/api/pith-number/IYYGEQYIKJZ547255DSSWREN7Z/graph.json","fetch_events":"https://pith.science/api/pith-number/IYYGEQYIKJZ547255DSSWREN7Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/action/storage_attestation","attest_author":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/action/author_attestation","sign_citation":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/action/citation_signature","submit_replication":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/action/replication_record"}},"created_at":"2026-05-25T02:01:21.619987+00:00","updated_at":"2026-05-25T02:01:21.619987+00:00"}