{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:INA3ZRYWGUTA6PUQHZ6SAOKHQD","short_pith_number":"pith:INA3ZRYW","schema_version":"1.0","canonical_sha256":"4341bcc71635260f3e903e7d20394780cfa0e88df256df3aa5350dab84d495b6","source":{"kind":"arxiv","id":"2310.16828","version":2},"attestation_state":"computed","paper":{"title":"TD-MPC2: Scalable, Robust World Models for Continuous Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters.","cross_cats":["cs.AI","cs.CV","cs.RO"],"primary_cat":"cs.LG","authors_text":"Hao Su, Nicklas Hansen, Xiaolong Wang","submitted_at":"2023-10-25T17:57:07Z","abstract_excerpt":"TD-MPC is a model-based reinforcement learning (RL) algorithm that performs local trajectory optimization in the latent space of a learned implicit (decoder-free) world model. In this work, we present TD-MPC2: a series of improvements upon the TD-MPC algorithm. We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perfo"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2310.16828","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:57:07Z","cross_cats_sorted":["cs.AI","cs.CV","cs.RO"],"title_canon_sha256":"f2ad3264774b571271338ae467cb30dc4224dc960e43751b19fe7de5d69db4b3","abstract_canon_sha256":"a2f737d999efdb1fbcc13897d2bbd2b8b944905f8270acce0ed123d3b92c7024"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:22.322284Z","signature_b64":"TkpNUHTQXM5za+sima/4M0S9Sg4Zc9Gjg3X/K3bsJPs7V9UPGyaSeE1zvtRlS7EgEWNzeTM3ASvUDfTnh9VWDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4341bcc71635260f3e903e7d20394780cfa0e88df256df3aa5350dab84d495b6","last_reissued_at":"2026-05-17T23:39:22.321510Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:22.321510Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TD-MPC2: Scalable, Robust World Models for Continuous Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters.","cross_cats":["cs.AI","cs.CV","cs.RO"],"primary_cat":"cs.LG","authors_text":"Hao Su, Nicklas Hansen, Xiaolong Wang","submitted_at":"2023-10-25T17:57:07Z","abstract_excerpt":"TD-MPC is a model-based reinforcement learning (RL) algorithm that performs local trajectory optimization in the latent space of a learned implicit (decoder-free) world model. In this work, we present TD-MPC2: a series of improvements upon the TD-MPC algorithm. We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perfo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perform 80 tasks across multiple task domains, embodiments, and action spaces.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The reported gains rely on the assumption that the chosen 104 tasks and four domains are representative enough that a single hyperparameter set will continue to work when the method is applied to new, unseen continuous-control problems.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"TD-MPC2 scales an implicit world-model RL method to a 317M-parameter agent that masters 80 tasks across four domains with a single hyperparameter configuration.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9d9457d575688ec95175cba1bfd515c0b36a0c0b76d352ed9c39c36f271ca75f"},"source":{"id":"2310.16828","kind":"arxiv","version":2},"verdict":{"id":"199b3dec-94d2-402a-9f97-e5c6f06258a3","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T17:21:09.097426Z","strongest_claim":"We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perform 80 tasks across multiple task domains, embodiments, and action spaces.","one_line_summary":"TD-MPC2 scales an implicit world-model RL method to a 317M-parameter agent that masters 80 tasks across four domains with a single hyperparameter configuration.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The reported gains rely on the assumption that the chosen 104 tasks and four domains are representative enough that a single hyperparameter set will continue to work when the method is applied to new, unseen continuous-control problems.","pith_extraction_headline":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters."},"references":{"count":162,"sample":[{"doi":"","year":2016,"title":"Layer normalization","work_id":"d34753e2-6f32-4f79-9705-ea5c146ebd8c","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Video pretraining (vpt): Learning to act by watching unlabeled online videos","work_id":"3ba8312e-03af-4ed5-8f67-041856cbe023","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"A distributional perspective on reinforcement learning","work_id":"b1fb027a-a8d8-4a70-bd30-47ca98dc5f84","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1957,"title":"A markovian decision process","work_id":"27354393-5813-4731-88c1-6372a303aefe","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1901,"title":"Language models are few-shot learners","work_id":"06905937-01c1-4635-ab7b-151e446701ad","ref_index":7,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":162,"snapshot_sha256":"9705adb397d177f7637798c09f0e4dec6fd0adebbbac081623d69202093275db","internal_anchors":12},"formal_canon":{"evidence_count":3,"snapshot_sha256":"5848add13880657af845b8fb0ab03a545150627f2fb09cf60d22024a54a666cf"},"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":"2310.16828","created_at":"2026-05-17T23:39:22.321637+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.16828v2","created_at":"2026-05-17T23:39:22.321637+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16828","created_at":"2026-05-17T23:39:22.321637+00:00"},{"alias_kind":"pith_short_12","alias_value":"INA3ZRYWGUTA","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"INA3ZRYWGUTA6PUQ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"INA3ZRYW","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":37,"internal_anchor_count":37,"sample":[{"citing_arxiv_id":"2506.14135","citing_title":"GAF: Gaussian Action Field as a 4D Representation for Dynamic World Modeling in Robotic Manipulation","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2510.03508","citing_title":"D2 Actor Critic: Diffusion Actor Meets Distributional Critic","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2310.02635","citing_title":"Reinforcement Learning with Foundation Priors: Let the Embodied Agent Efficiently Learn on Its Own","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2502.05907","citing_title":"EvolvingAgent: Curriculum Self-evolving Agent with Continual World Model for Long-Horizon Tasks","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21800","citing_title":"stable-worldmodel: A Platform for Reproducible World Modeling Research and Evaluation","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16398","citing_title":"Support-Safe Variational Hybrid Filtering for Contact-Mode and Sparse-Law Recovery","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16725","citing_title":"Baba in Wonderland: Online Self-Supervised Dynamics Discovery for Executable World Models","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2506.05762","citing_title":"BiTrajDiff: Bidirectional Trajectory Generation with Diffusion Models for Offline Reinforcement Learning","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2509.20869","citing_title":"Model-Based Reinforcement Learning under Random Observation Delays","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2411.04983","citing_title":"DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2512.17091","citing_title":"Learning to Plan, Planning to Learn: Adaptive Hierarchical RL-MPC for Sample-Efficient Decision Making","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2602.11075","citing_title":"RISE: Self-Improving Robot Policy with Compositional World Model","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"2603.07083","citing_title":"Dreamer-CDP: Improving Reconstruction-free World Models Via Continuous Deterministic Representation Prediction","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2603.09030","citing_title":"PlayWorld: Learning Robot World Models from Autonomous Play","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2509.24527","citing_title":"Training Agents Inside of Scalable World Models","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13013","citing_title":"JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03208","citing_title":"Hierarchical Planning with Latent World Models","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02911","citing_title":"Learning Task-Invariant Properties via Dreamer: Enabling Efficient Policy Transfer for Quadruped Robots","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03449","citing_title":"Neural Operators for Multi-Task Control and Adaptation","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11473","citing_title":"TOPPO: Rethinking PPO for Multi-Task Reinforcement Learning with Critic Balancing","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2601.16163","citing_title":"Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27450","citing_title":"RAY-TOLD: Ray-Based Latent Dynamics for Dense Dynamic Obstacle Avoidance with TDMPC","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27411","citing_title":"Detecting is Easy, Adapting is Hard: Local Expert Growth for Visual Model-Based Reinforcement Learning under Distribution Shift","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09364","citing_title":"Multi-scale Predictive Representations for Goal-conditioned Reinforcement Learning","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08954","citing_title":"MolWorld: Molecule World Models for Actionable Molecular Optimization","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":3,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD","json":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD.json","graph_json":"https://pith.science/api/pith-number/INA3ZRYWGUTA6PUQHZ6SAOKHQD/graph.json","events_json":"https://pith.science/api/pith-number/INA3ZRYWGUTA6PUQHZ6SAOKHQD/events.json","paper":"https://pith.science/paper/INA3ZRYW"},"agent_actions":{"view_html":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD","download_json":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD.json","view_paper":"https://pith.science/paper/INA3ZRYW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.16828&json=true","fetch_graph":"https://pith.science/api/pith-number/INA3ZRYWGUTA6PUQHZ6SAOKHQD/graph.json","fetch_events":"https://pith.science/api/pith-number/INA3ZRYWGUTA6PUQHZ6SAOKHQD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD/action/storage_attestation","attest_author":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD/action/author_attestation","sign_citation":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD/action/citation_signature","submit_replication":"https://pith.science/pith/INA3ZRYWGUTA6PUQHZ6SAOKHQD/action/replication_record"}},"created_at":"2026-05-17T23:39:22.321637+00:00","updated_at":"2026-05-17T23:39:22.321637+00:00"}