{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MBW3LIJ5GYXQBUZJKTORKI7BVF","short_pith_number":"pith:MBW3LIJ5","schema_version":"1.0","canonical_sha256":"606db5a13d362f00d32954dd1523e1a946a85e20f47b6bb81fd54c6e0ea768e2","source":{"kind":"arxiv","id":"2506.21782","version":1},"attestation_state":"computed","paper":{"title":"M3PO: Massively Multi-Task Model-Based Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Aditya Narendra, Aleksandr Panov, Dmitry Makarov","submitted_at":"2025-06-26T21:39:01Z","abstract_excerpt":"We introduce Massively Multi-Task Model-Based Policy Optimization (M3PO), a scalable model-based reinforcement learning (MBRL) framework designed to address sample inefficiency in single-task settings and poor generalization in multi-task domains. Existing model-based approaches like DreamerV3 rely on pixel-level generative models that neglect control-centric representations, while model-free methods such as PPO suffer from high sample complexity and weak exploration. M3PO integrates an implicit world model, trained to predict task outcomes without observation reconstruction, with a hybrid exp"},"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":"2506.21782","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-26T21:39:01Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"18bb3269119cd5dd99c626764b993459cbc4b7bcf564d4f00a96c7c68e254c61","abstract_canon_sha256":"8f892230db0242756ddbbc82572ff43b185997ee966e0832aeb213e42dbf64cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:28:00.975523Z","signature_b64":"bqjdA3/xKSKwklHjf3hGKDysFQQa25zgqqDNPv/i38lW/cZn/6VPJDXY9Mt8wsegTS1Z4u7efH+4+jdqv4W/Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"606db5a13d362f00d32954dd1523e1a946a85e20f47b6bb81fd54c6e0ea768e2","last_reissued_at":"2026-07-05T11:28:00.975027Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:28:00.975027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M3PO: Massively Multi-Task Model-Based Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Aditya Narendra, Aleksandr Panov, Dmitry Makarov","submitted_at":"2025-06-26T21:39:01Z","abstract_excerpt":"We introduce Massively Multi-Task Model-Based Policy Optimization (M3PO), a scalable model-based reinforcement learning (MBRL) framework designed to address sample inefficiency in single-task settings and poor generalization in multi-task domains. Existing model-based approaches like DreamerV3 rely on pixel-level generative models that neglect control-centric representations, while model-free methods such as PPO suffer from high sample complexity and weak exploration. M3PO integrates an implicit world model, trained to predict task outcomes without observation reconstruction, with a hybrid exp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.21782","kind":"arxiv","version":1},"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/2506.21782/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":"2506.21782","created_at":"2026-07-05T11:28:00.975082+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.21782v1","created_at":"2026-07-05T11:28:00.975082+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.21782","created_at":"2026-07-05T11:28:00.975082+00:00"},{"alias_kind":"pith_short_12","alias_value":"MBW3LIJ5GYXQ","created_at":"2026-07-05T11:28:00.975082+00:00"},{"alias_kind":"pith_short_16","alias_value":"MBW3LIJ5GYXQBUZJ","created_at":"2026-07-05T11:28:00.975082+00:00"},{"alias_kind":"pith_short_8","alias_value":"MBW3LIJ5","created_at":"2026-07-05T11:28:00.975082+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/MBW3LIJ5GYXQBUZJKTORKI7BVF","json":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF.json","graph_json":"https://pith.science/api/pith-number/MBW3LIJ5GYXQBUZJKTORKI7BVF/graph.json","events_json":"https://pith.science/api/pith-number/MBW3LIJ5GYXQBUZJKTORKI7BVF/events.json","paper":"https://pith.science/paper/MBW3LIJ5"},"agent_actions":{"view_html":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF","download_json":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF.json","view_paper":"https://pith.science/paper/MBW3LIJ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.21782&json=true","fetch_graph":"https://pith.science/api/pith-number/MBW3LIJ5GYXQBUZJKTORKI7BVF/graph.json","fetch_events":"https://pith.science/api/pith-number/MBW3LIJ5GYXQBUZJKTORKI7BVF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF/action/storage_attestation","attest_author":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF/action/author_attestation","sign_citation":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF/action/citation_signature","submit_replication":"https://pith.science/pith/MBW3LIJ5GYXQBUZJKTORKI7BVF/action/replication_record"}},"created_at":"2026-07-05T11:28:00.975082+00:00","updated_at":"2026-07-05T11:28:00.975082+00:00"}