{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:W3XM67W7Z52BXDXZG3Q7YNRIA7","short_pith_number":"pith:W3XM67W7","schema_version":"1.0","canonical_sha256":"b6eecf7edfcf741b8ef936e1fc362807efadebe83f6fe1f2abd49d308f4bcedf","source":{"kind":"arxiv","id":"2410.18076","version":4},"attestation_state":"computed","paper":{"title":"Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kevin Frans, Max Wilcoxson, Qiyang Li, Sergey Levine","submitted_at":"2024-10-23T17:58:45Z","abstract_excerpt":"Unsupervised pretraining has been transformative in many supervised domains. However, applying such ideas to reinforcement learning (RL) presents a unique challenge in that fine-tuning does not involve mimicking task-specific data, but rather exploring and locating the solution through iterative self-improvement. In this work, we study how unlabeled offline trajectory data can be leveraged to learn efficient exploration strategies. While prior data can be used to pretrain a set of low-level skills, or as additional off-policy data for online RL, it has been unclear how to combine these ideas e"},"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":"2410.18076","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T17:58:45Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"64c5b308469caac12989855236599954cde746fe6913b22a7933657389a6b7df","abstract_canon_sha256":"9d814e7747f1bf4055a73deff85349184dee5431a18893d785db85d3119577a8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:51.731895Z","signature_b64":"huLZQtM7I9U6Xk5y3rpUEbt0xHLjRKiARG0RpXw6RBkGvDbwt1YZdC8G3HvfU553ewDtg0/WTVo5zZ7iXD47Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b6eecf7edfcf741b8ef936e1fc362807efadebe83f6fe1f2abd49d308f4bcedf","last_reissued_at":"2026-07-05T11:35:51.731446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:51.731446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kevin Frans, Max Wilcoxson, Qiyang Li, Sergey Levine","submitted_at":"2024-10-23T17:58:45Z","abstract_excerpt":"Unsupervised pretraining has been transformative in many supervised domains. However, applying such ideas to reinforcement learning (RL) presents a unique challenge in that fine-tuning does not involve mimicking task-specific data, but rather exploring and locating the solution through iterative self-improvement. In this work, we study how unlabeled offline trajectory data can be leveraged to learn efficient exploration strategies. While prior data can be used to pretrain a set of low-level skills, or as additional off-policy data for online RL, it has been unclear how to combine these ideas e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.18076","kind":"arxiv","version":4},"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/2410.18076/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":"2410.18076","created_at":"2026-07-05T11:35:51.731503+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.18076v4","created_at":"2026-07-05T11:35:51.731503+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.18076","created_at":"2026-07-05T11:35:51.731503+00:00"},{"alias_kind":"pith_short_12","alias_value":"W3XM67W7Z52B","created_at":"2026-07-05T11:35:51.731503+00:00"},{"alias_kind":"pith_short_16","alias_value":"W3XM67W7Z52BXDXZ","created_at":"2026-07-05T11:35:51.731503+00:00"},{"alias_kind":"pith_short_8","alias_value":"W3XM67W7","created_at":"2026-07-05T11:35:51.731503+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31958","citing_title":"Adapting Generalist Robot Policies with Semantic Reinforcement Learning","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2507.07969","citing_title":"Reinforcement Learning with Action Chunking","ref_index":82,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7","json":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7.json","graph_json":"https://pith.science/api/pith-number/W3XM67W7Z52BXDXZG3Q7YNRIA7/graph.json","events_json":"https://pith.science/api/pith-number/W3XM67W7Z52BXDXZG3Q7YNRIA7/events.json","paper":"https://pith.science/paper/W3XM67W7"},"agent_actions":{"view_html":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7","download_json":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7.json","view_paper":"https://pith.science/paper/W3XM67W7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.18076&json=true","fetch_graph":"https://pith.science/api/pith-number/W3XM67W7Z52BXDXZG3Q7YNRIA7/graph.json","fetch_events":"https://pith.science/api/pith-number/W3XM67W7Z52BXDXZG3Q7YNRIA7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7/action/storage_attestation","attest_author":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7/action/author_attestation","sign_citation":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7/action/citation_signature","submit_replication":"https://pith.science/pith/W3XM67W7Z52BXDXZG3Q7YNRIA7/action/replication_record"}},"created_at":"2026-07-05T11:35:51.731503+00:00","updated_at":"2026-07-05T11:35:51.731503+00:00"}