{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:IYYGEQYIKJZ547255DSSWREN7Z","short_pith_number":"pith:IYYGEQYI","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"},"canonical_sha256":"46306243085273de7f5de8e52b448dfe7dcaee771d01773e0800d44bd945530b","source":{"kind":"arxiv","id":"2605.04568","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.04568","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"arxiv_version","alias_value":"2605.04568v2","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04568","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"pith_short_12","alias_value":"IYYGEQYIKJZ5","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"pith_short_16","alias_value":"IYYGEQYIKJZ54725","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"pith_short_8","alias_value":"IYYGEQYI","created_at":"2026-05-25T02:01:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:IYYGEQYIKJZ547255DSSWREN7Z","target":"record","payload":{"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"},"canonical_sha256":"46306243085273de7f5de8e52b448dfe7dcaee771d01773e0800d44bd945530b","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"},"source_kind":"arxiv","source_id":"2605.04568","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-25T02:01:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JE1VVup5T2epXf5lyweuKfI9jbyHFx4SEEcTxEf3zRX4sv8/DbUuUVAhO8OMBXbI+S+Q9ZYwKR8suqdfvX3oBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T08:19:55.089439Z"},"content_sha256":"c61a9a5b38af6515a98a922eebca4f37a2a11678b39f59afeaf2565192d5427d","schema_version":"1.0","event_id":"sha256:c61a9a5b38af6515a98a922eebca4f37a2a11678b39f59afeaf2565192d5427d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:IYYGEQYIKJZ547255DSSWREN7Z","target":"graph","payload":{"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"},"verdict_id":"3e10bde3-1524-4c77-a33e-565509147905"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-25T02:01:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KYxeVWsM9H4tpIWT7zGDKml4xYtLEFrEhxN2GbK/OQPXukImLn2GwPUR7DVU5FWU4XgQdAZbB4wsb9PqFIe9CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T08:19:55.089914Z"},"content_sha256":"72588de5bbf7424d5a7aea2a0e79da9de82287e6f44a250164d206bed3c0e583","schema_version":"1.0","event_id":"sha256:72588de5bbf7424d5a7aea2a0e79da9de82287e6f44a250164d206bed3c0e583"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/bundle.json","state_url":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IYYGEQYIKJZ547255DSSWREN7Z/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-17T08:19:55Z","links":{"resolver":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z","bundle":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/bundle.json","state":"https://pith.science/pith/IYYGEQYIKJZ547255DSSWREN7Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IYYGEQYIKJZ547255DSSWREN7Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IYYGEQYIKJZ547255DSSWREN7Z","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b601aed29fbd7d25fadca9352d056cdb7c42af1fe561f93e9ca9cf19c2980a09","cross_cats_sorted":["cs.AI","cs.RO"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-06T07:13:11Z","title_canon_sha256":"10edec659906056595f8435feb3dabf1501bab40dc053cbdf663f40b8cd86fbc"},"schema_version":"1.0","source":{"id":"2605.04568","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.04568","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"arxiv_version","alias_value":"2605.04568v2","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04568","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"pith_short_12","alias_value":"IYYGEQYIKJZ5","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"pith_short_16","alias_value":"IYYGEQYIKJZ54725","created_at":"2026-05-25T02:01:21Z"},{"alias_kind":"pith_short_8","alias_value":"IYYGEQYI","created_at":"2026-05-25T02:01:21Z"}],"graph_snapshots":[{"event_id":"sha256:72588de5bbf7424d5a7aea2a0e79da9de82287e6f44a250164d206bed3c0e583","target":"graph","created_at":"2026-05-25T02:01:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance."}],"snapshot_sha256":"12fdb68e98465d566cdab351d22705ffba0aa13bc225bbcf49e8ca24164555df"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2605.04568/integrity.json","findings":[],"snapshot_sha256":"7a5f27b019c08312cefe236fc31addcf375473790e0ef05a1917df7b1cec37de","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Jonathan Spieler, Sven Behnke","cross_cats":["cs.AI","cs.RO"],"headline":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance.","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-06T07:13:11Z","title":"Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.04568","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T16:27:37.476846Z","id":"3e10bde3-1524-4c77-a33e-565509147905","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"Dream-MPC refines a few policy-rollout trajectories via gradient ascent inside a learned world model to raise overall task performance.","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.","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."}},"verdict_id":"3e10bde3-1524-4c77-a33e-565509147905"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c61a9a5b38af6515a98a922eebca4f37a2a11678b39f59afeaf2565192d5427d","target":"record","created_at":"2026-05-25T02:01:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"b601aed29fbd7d25fadca9352d056cdb7c42af1fe561f93e9ca9cf19c2980a09","cross_cats_sorted":["cs.AI","cs.RO"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-06T07:13:11Z","title_canon_sha256":"10edec659906056595f8435feb3dabf1501bab40dc053cbdf663f40b8cd86fbc"},"schema_version":"1.0","source":{"id":"2605.04568","kind":"arxiv","version":2}},"canonical_sha256":"46306243085273de7f5de8e52b448dfe7dcaee771d01773e0800d44bd945530b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"46306243085273de7f5de8e52b448dfe7dcaee771d01773e0800d44bd945530b","first_computed_at":"2026-05-25T02:01:21.619872Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-25T02:01:21.619872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wegpn6Frnal9Jau6WP8K9EPBgwiC1A+THdiFF05svJXVJLUztfPyy2SgQq2CWf0RFUaS6/ogiphU+qkbpfJKAQ==","signature_status":"signed_v1","signed_at":"2026-05-25T02:01:21.620637Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.04568","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c61a9a5b38af6515a98a922eebca4f37a2a11678b39f59afeaf2565192d5427d","sha256:72588de5bbf7424d5a7aea2a0e79da9de82287e6f44a250164d206bed3c0e583"],"state_sha256":"3354e70da8d60ebf80283abd79490b337bed50b24cfe0a137cd9732e31320701"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bYySU/nXhiH/kRKpU84BOOllaTQ5z4UukqASSC3ONAFDWMoZIz9qtqEoO47xzpVUenaFy851WtIJuILTlcyhAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-17T08:19:55.092571Z","bundle_sha256":"c2916f4ce5c4cda61f52674c6a62ea4dbd37f2ba3ab5075636a5502b3c8db606"}}