{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:KOHZSHTUP4UHNLJOR5KZT4VMWE","short_pith_number":"pith:KOHZSHTU","schema_version":"1.0","canonical_sha256":"538f991e747f2876ad2e8f5599f2acb12733c406fc60f9e5678fe7a5521f4850","source":{"kind":"arxiv","id":"2205.08129","version":2},"attestation_state":"computed","paper":{"title":"Planning to Practice: Efficient Online Fine-Tuning by Composing Goals in Latent Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Ashvin Nair, Kuan Fang, Patrick Yin, Sergey Levine","submitted_at":"2022-05-17T06:58:17Z","abstract_excerpt":"General-purpose robots require diverse repertoires of behaviors to complete challenging tasks in real-world unstructured environments. To address this issue, goal-conditioned reinforcement learning aims to acquire policies that can reach configurable goals for a wide range of tasks on command. However, such goal-conditioned policies are notoriously difficult and time-consuming to train from scratch. In this paper, we propose Planning to Practice (PTP), a method that makes it practical to train goal-conditioned policies for long-horizon tasks that require multiple distinct types of interactions"},"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":"2205.08129","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2022-05-17T06:58:17Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"2d8db0cb7db3a7159b258d912de2823bf65a3416ffb4cc2871120415f353bad9","abstract_canon_sha256":"70ddca29045a59d52c450eff27d13d02d5fab77d9dcc395c5e42ae84f99f5831"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:01:57.058250Z","signature_b64":"k65+GWcLKUO+qKiLUFBqbu1oLFfr6PS6Aw0ps2lyySG/tyI8G/mw7IgZPE+ycd+VyDcyZbN5kn4GXsi1ahdbCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"538f991e747f2876ad2e8f5599f2acb12733c406fc60f9e5678fe7a5521f4850","last_reissued_at":"2026-07-05T06:01:57.057771Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:01:57.057771Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Planning to Practice: Efficient Online Fine-Tuning by Composing Goals in Latent Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Ashvin Nair, Kuan Fang, Patrick Yin, Sergey Levine","submitted_at":"2022-05-17T06:58:17Z","abstract_excerpt":"General-purpose robots require diverse repertoires of behaviors to complete challenging tasks in real-world unstructured environments. To address this issue, goal-conditioned reinforcement learning aims to acquire policies that can reach configurable goals for a wide range of tasks on command. However, such goal-conditioned policies are notoriously difficult and time-consuming to train from scratch. In this paper, we propose Planning to Practice (PTP), a method that makes it practical to train goal-conditioned policies for long-horizon tasks that require multiple distinct types of interactions"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.08129","kind":"arxiv","version":2},"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/2205.08129/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":"2205.08129","created_at":"2026-07-05T06:01:57.057831+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.08129v2","created_at":"2026-07-05T06:01:57.057831+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.08129","created_at":"2026-07-05T06:01:57.057831+00:00"},{"alias_kind":"pith_short_12","alias_value":"KOHZSHTUP4UH","created_at":"2026-07-05T06:01:57.057831+00:00"},{"alias_kind":"pith_short_16","alias_value":"KOHZSHTUP4UHNLJO","created_at":"2026-07-05T06:01:57.057831+00:00"},{"alias_kind":"pith_short_8","alias_value":"KOHZSHTU","created_at":"2026-07-05T06:01:57.057831+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":109,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE","json":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE.json","graph_json":"https://pith.science/api/pith-number/KOHZSHTUP4UHNLJOR5KZT4VMWE/graph.json","events_json":"https://pith.science/api/pith-number/KOHZSHTUP4UHNLJOR5KZT4VMWE/events.json","paper":"https://pith.science/paper/KOHZSHTU"},"agent_actions":{"view_html":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE","download_json":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE.json","view_paper":"https://pith.science/paper/KOHZSHTU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.08129&json=true","fetch_graph":"https://pith.science/api/pith-number/KOHZSHTUP4UHNLJOR5KZT4VMWE/graph.json","fetch_events":"https://pith.science/api/pith-number/KOHZSHTUP4UHNLJOR5KZT4VMWE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE/action/storage_attestation","attest_author":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE/action/author_attestation","sign_citation":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE/action/citation_signature","submit_replication":"https://pith.science/pith/KOHZSHTUP4UHNLJOR5KZT4VMWE/action/replication_record"}},"created_at":"2026-07-05T06:01:57.057831+00:00","updated_at":"2026-07-05T06:01:57.057831+00:00"}