{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:INA3ZRYWGUTA6PUQHZ6SAOKHQD","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":"a2f737d999efdb1fbcc13897d2bbd2b8b944905f8270acce0ed123d3b92c7024","cross_cats_sorted":["cs.AI","cs.CV","cs.RO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:57:07Z","title_canon_sha256":"f2ad3264774b571271338ae467cb30dc4224dc960e43751b19fe7de5d69db4b3"},"schema_version":"1.0","source":{"id":"2310.16828","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.16828","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"arxiv_version","alias_value":"2310.16828v2","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16828","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"pith_short_12","alias_value":"INA3ZRYWGUTA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"INA3ZRYWGUTA6PUQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"INA3ZRYW","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:082fd3addb6655a5e4bf087f7e1bacaad7dc2462f65ca1550cdace617df8509b","target":"graph","created_at":"2026-05-17T23:39:22Z","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":"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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters."}],"snapshot_sha256":"9d9457d575688ec95175cba1bfd515c0b36a0c0b76d352ed9c39c36f271ca75f"},"formal_canon":{"evidence_count":3,"snapshot_sha256":"5848add13880657af845b8fb0ab03a545150627f2fb09cf60d22024a54a666cf"},"paper":{"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","authors_text":"Hao Su, Nicklas Hansen, Xiaolong Wang","cross_cats":["cs.AI","cs.CV","cs.RO"],"headline":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:57:07Z","title":"TD-MPC2: Scalable, Robust World Models for Continuous Control"},"references":{"count":162,"internal_anchors":12,"resolved_work":162,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Layer normalization","work_id":"d34753e2-6f32-4f79-9705-ea5c146ebd8c","year":2016},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Video pretraining (vpt): Learning to act by watching unlabeled online videos","work_id":"3ba8312e-03af-4ed5-8f67-041856cbe023","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"A distributional perspective on reinforcement learning","work_id":"b1fb027a-a8d8-4a70-bd30-47ca98dc5f84","year":2017},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"A markovian decision process","work_id":"27354393-5813-4731-88c1-6372a303aefe","year":1957},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":7,"title":"Language models are few-shot learners","work_id":"06905937-01c1-4635-ab7b-151e446701ad","year":1901}],"snapshot_sha256":"9705adb397d177f7637798c09f0e4dec6fd0adebbbac081623d69202093275db"},"source":{"id":"2310.16828","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-14T17:21:09.097426Z","id":"199b3dec-94d2-402a-9f97-e5c6f06258a3","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"TD-MPC2 achieves significantly better performance than baselines on 104 continuous control tasks using one fixed set of hyperparameters.","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.","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."}},"verdict_id":"199b3dec-94d2-402a-9f97-e5c6f06258a3"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e9e6e9783524fa6a9cff95e0ce591eaf411642763ad82859a025ce7bec590292","target":"record","created_at":"2026-05-17T23:39:22Z","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":"a2f737d999efdb1fbcc13897d2bbd2b8b944905f8270acce0ed123d3b92c7024","cross_cats_sorted":["cs.AI","cs.CV","cs.RO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:57:07Z","title_canon_sha256":"f2ad3264774b571271338ae467cb30dc4224dc960e43751b19fe7de5d69db4b3"},"schema_version":"1.0","source":{"id":"2310.16828","kind":"arxiv","version":2}},"canonical_sha256":"4341bcc71635260f3e903e7d20394780cfa0e88df256df3aa5350dab84d495b6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4341bcc71635260f3e903e7d20394780cfa0e88df256df3aa5350dab84d495b6","first_computed_at":"2026-05-17T23:39:22.321510Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:22.321510Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TkpNUHTQXM5za+sima/4M0S9Sg4Zc9Gjg3X/K3bsJPs7V9UPGyaSeE1zvtRlS7EgEWNzeTM3ASvUDfTnh9VWDw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:22.322284Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.16828","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e9e6e9783524fa6a9cff95e0ce591eaf411642763ad82859a025ce7bec590292","sha256:082fd3addb6655a5e4bf087f7e1bacaad7dc2462f65ca1550cdace617df8509b"],"state_sha256":"7df06bb34980fa68279a14783daf426d233971f6aad9049e47ebb9f9ce18421a"}