{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:AC6JJ5VUKACLWMNDABSYFXKZLX","short_pith_number":"pith:AC6JJ5VU","schema_version":"1.0","canonical_sha256":"00bc94f6b45004bb31a3006582dd595dcca34425a9f1785fd24b310d9ed30c43","source":{"kind":"arxiv","id":"2110.15191","version":1},"attestation_state":"computed","paper":{"title":"URLB: Unsupervised Reinforcement Learning Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Albert Zhan, Catherine Cang, Denis Yarats, Hao Liu, Kevin Lu, Kimin Lee, Lerrel Pinto, Michael Laskin, Pieter Abbeel","submitted_at":"2021-10-28T15:07:01Z","abstract_excerpt":"Deep Reinforcement Learning (RL) has emerged as a powerful paradigm to solve a range of complex yet specific control tasks. Yet training generalist agents that can quickly adapt to new tasks remains an outstanding challenge. Recent advances in unsupervised RL have shown that pre-training RL agents with self-supervised intrinsic rewards can result in efficient adaptation. However, these algorithms have been hard to compare and develop due to the lack of a unified benchmark. To this end, we introduce the Unsupervised Reinforcement Learning Benchmark (URLB). URLB consists of two phases: reward-fr"},"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":"2110.15191","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-10-28T15:07:01Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"c33f1e98f57899bd4cfee5365eb2604cfe065926395d527d1496c7c6b7073672","abstract_canon_sha256":"808e7a75837b8b35e50b19280cd66a3412bae8bdc12757049bfe97d9a18c9e5e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:26:52.554113Z","signature_b64":"zC9pkxwgYJdWUcbTtF4t/yb2ocK+mqTu974lJVb09qVGvBzlLw+7yD9fkWAQAV7OiE9q0gvV1bjd+1n3Le4FDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00bc94f6b45004bb31a3006582dd595dcca34425a9f1785fd24b310d9ed30c43","last_reissued_at":"2026-07-05T03:26:52.553708Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:26:52.553708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"URLB: Unsupervised Reinforcement Learning Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Albert Zhan, Catherine Cang, Denis Yarats, Hao Liu, Kevin Lu, Kimin Lee, Lerrel Pinto, Michael Laskin, Pieter Abbeel","submitted_at":"2021-10-28T15:07:01Z","abstract_excerpt":"Deep Reinforcement Learning (RL) has emerged as a powerful paradigm to solve a range of complex yet specific control tasks. Yet training generalist agents that can quickly adapt to new tasks remains an outstanding challenge. Recent advances in unsupervised RL have shown that pre-training RL agents with self-supervised intrinsic rewards can result in efficient adaptation. However, these algorithms have been hard to compare and develop due to the lack of a unified benchmark. To this end, we introduce the Unsupervised Reinforcement Learning Benchmark (URLB). URLB consists of two phases: reward-fr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.15191","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/2110.15191/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":"2110.15191","created_at":"2026-07-05T03:26:52.553767+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.15191v1","created_at":"2026-07-05T03:26:52.553767+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.15191","created_at":"2026-07-05T03:26:52.553767+00:00"},{"alias_kind":"pith_short_12","alias_value":"AC6JJ5VUKACL","created_at":"2026-07-05T03:26:52.553767+00:00"},{"alias_kind":"pith_short_16","alias_value":"AC6JJ5VUKACLWMND","created_at":"2026-07-05T03:26:52.553767+00:00"},{"alias_kind":"pith_short_8","alias_value":"AC6JJ5VU","created_at":"2026-07-05T03:26:52.553767+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21271","citing_title":"Reward-free Pretraining for Reinforcement Learning via Occupancy Coverage Maximization","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2412.08812","citing_title":"Test-Time Alignment via Hypothesis Reweighting","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20209","citing_title":"NaP-Control: Navigating Diffusion Prior for Versatile and Fast Character Control","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2310.01889","citing_title":"Ring Attention with Blockwise Transformers for Near-Infinite Context","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX","json":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX.json","graph_json":"https://pith.science/api/pith-number/AC6JJ5VUKACLWMNDABSYFXKZLX/graph.json","events_json":"https://pith.science/api/pith-number/AC6JJ5VUKACLWMNDABSYFXKZLX/events.json","paper":"https://pith.science/paper/AC6JJ5VU"},"agent_actions":{"view_html":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX","download_json":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX.json","view_paper":"https://pith.science/paper/AC6JJ5VU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.15191&json=true","fetch_graph":"https://pith.science/api/pith-number/AC6JJ5VUKACLWMNDABSYFXKZLX/graph.json","fetch_events":"https://pith.science/api/pith-number/AC6JJ5VUKACLWMNDABSYFXKZLX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX/action/storage_attestation","attest_author":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX/action/author_attestation","sign_citation":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX/action/citation_signature","submit_replication":"https://pith.science/pith/AC6JJ5VUKACLWMNDABSYFXKZLX/action/replication_record"}},"created_at":"2026-07-05T03:26:52.553767+00:00","updated_at":"2026-07-05T03:26:52.553767+00:00"}