{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:RED7FTH27HHHHNUT2LQL2UCCFT","short_pith_number":"pith:RED7FTH2","schema_version":"1.0","canonical_sha256":"8907f2ccfaf9ce73b693d2e0bd50422cf290a7b1e5ff80a613b75774e3a7600c","source":{"kind":"arxiv","id":"2208.10291","version":3},"attestation_state":"computed","paper":{"title":"Efficient Planning in a Compact Latent Action Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Edward Grefenstette, Michael Janner, Tianjun Zhang, Tim Rockt\\\"aschel, Yuandong Tian, Yueying Li, Zhengyao Jiang","submitted_at":"2022-08-22T13:19:02Z","abstract_excerpt":"Planning-based reinforcement learning has shown strong performance in tasks in discrete and low-dimensional continuous action spaces. However, planning usually brings significant computational overhead for decision-making, and scaling such methods to high-dimensional action spaces remains challenging. To advance efficient planning for high-dimensional continuous control, we propose Trajectory Autoencoding Planner (TAP), which learns low-dimensional latent action codes with a state-conditional VQ-VAE. The decoder of the VQ-VAE thus serves as a novel dynamics model that takes latent actions and "},"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":"2208.10291","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-22T13:19:02Z","cross_cats_sorted":[],"title_canon_sha256":"4cb900d91dea91c735a0aca7c3d7f762978761c4e7fecde2d02797e5dcedd11c","abstract_canon_sha256":"554ec622b2f5dd859e3a47ac978e277f95d488acb02fc8aaee612d3f4185be8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:35:20.143928Z","signature_b64":"fQazVNAPS2efQwoHmrN8gXeM7RITTIcDwhyjNK5A4RJQrFNu/0NcnKqw1GoItW53pTjlYST1jGdM6BzqqLmaBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8907f2ccfaf9ce73b693d2e0bd50422cf290a7b1e5ff80a613b75774e3a7600c","last_reissued_at":"2026-07-05T05:35:20.143407Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:35:20.143407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Planning in a Compact Latent Action Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Edward Grefenstette, Michael Janner, Tianjun Zhang, Tim Rockt\\\"aschel, Yuandong Tian, Yueying Li, Zhengyao Jiang","submitted_at":"2022-08-22T13:19:02Z","abstract_excerpt":"Planning-based reinforcement learning has shown strong performance in tasks in discrete and low-dimensional continuous action spaces. However, planning usually brings significant computational overhead for decision-making, and scaling such methods to high-dimensional action spaces remains challenging. To advance efficient planning for high-dimensional continuous control, we propose Trajectory Autoencoding Planner (TAP), which learns low-dimensional latent action codes with a state-conditional VQ-VAE. The decoder of the VQ-VAE thus serves as a novel dynamics model that takes latent actions and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.10291","kind":"arxiv","version":3},"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/2208.10291/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":"2208.10291","created_at":"2026-07-05T05:35:20.143479+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.10291v3","created_at":"2026-07-05T05:35:20.143479+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.10291","created_at":"2026-07-05T05:35:20.143479+00:00"},{"alias_kind":"pith_short_12","alias_value":"RED7FTH27HHH","created_at":"2026-07-05T05:35:20.143479+00:00"},{"alias_kind":"pith_short_16","alias_value":"RED7FTH27HHHHNUT","created_at":"2026-07-05T05:35:20.143479+00:00"},{"alias_kind":"pith_short_8","alias_value":"RED7FTH2","created_at":"2026-07-05T05:35:20.143479+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22480","citing_title":"ARP: Enhancing Quantized Skill Abstractions via Visual Alignment and Iterative Refinement for Robotic Manipulation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2511.18085","citing_title":"Continually Evolving Skill Knowledge in Vision Language Action Model","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":162,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT","json":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT.json","graph_json":"https://pith.science/api/pith-number/RED7FTH27HHHHNUT2LQL2UCCFT/graph.json","events_json":"https://pith.science/api/pith-number/RED7FTH27HHHHNUT2LQL2UCCFT/events.json","paper":"https://pith.science/paper/RED7FTH2"},"agent_actions":{"view_html":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT","download_json":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT.json","view_paper":"https://pith.science/paper/RED7FTH2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.10291&json=true","fetch_graph":"https://pith.science/api/pith-number/RED7FTH27HHHHNUT2LQL2UCCFT/graph.json","fetch_events":"https://pith.science/api/pith-number/RED7FTH27HHHHNUT2LQL2UCCFT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT/action/storage_attestation","attest_author":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT/action/author_attestation","sign_citation":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT/action/citation_signature","submit_replication":"https://pith.science/pith/RED7FTH27HHHHNUT2LQL2UCCFT/action/replication_record"}},"created_at":"2026-07-05T05:35:20.143479+00:00","updated_at":"2026-07-05T05:35:20.143479+00:00"}