{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:UFTTKZ4GK7L3BZGJMUGAB6RJHT","short_pith_number":"pith:UFTTKZ4G","schema_version":"1.0","canonical_sha256":"a16735678657d7b0e4c9650c00fa293ceec8b9f6f68ca8d9dbd74ad91df9a816","source":{"kind":"arxiv","id":"2310.03718","version":2},"attestation_state":"computed","paper":{"title":"Constraint-Conditioned Policy Optimization for Versatile Safe Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ding Zhao, Jiacheng Zhu, Tingnan Zhang, Wenhao Yu, Yihang Yao, Zhepeng Cen, Zuxin Liu","submitted_at":"2023-10-05T17:39:02Z","abstract_excerpt":"Safe reinforcement learning (RL) focuses on training reward-maximizing agents subject to pre-defined safety constraints. Yet, learning versatile safe policies that can adapt to varying safety constraint requirements during deployment without retraining remains a largely unexplored and challenging area. In this work, we formulate the versatile safe RL problem and consider two primary requirements: training efficiency and zero-shot adaptation capability. To address them, we introduce the Conditioned Constrained Policy Optimization (CCPO) framework, consisting of two key modules: (1) Versatile Va"},"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":"2310.03718","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-05T17:39:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7fdaf1e06ce39accaa5805bca317d0d2950aa9e8f2ed33b044a66213317ebcc3","abstract_canon_sha256":"3980100c08ca7e0e70c8584247c5737a4c3beb56ad498ed4c3e9e051497ab98d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:13:27.360633Z","signature_b64":"2fpkw+ntVqNr+d072VJX3/BtcqpJUTXHK5FUxHuE5IjZKlxVJSJy0gkdWhPZpVGpanlW8HqU9wm6XKKzhISnBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a16735678657d7b0e4c9650c00fa293ceec8b9f6f68ca8d9dbd74ad91df9a816","last_reissued_at":"2026-07-05T08:13:27.360094Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:13:27.360094Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Constraint-Conditioned Policy Optimization for Versatile Safe Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ding Zhao, Jiacheng Zhu, Tingnan Zhang, Wenhao Yu, Yihang Yao, Zhepeng Cen, Zuxin Liu","submitted_at":"2023-10-05T17:39:02Z","abstract_excerpt":"Safe reinforcement learning (RL) focuses on training reward-maximizing agents subject to pre-defined safety constraints. Yet, learning versatile safe policies that can adapt to varying safety constraint requirements during deployment without retraining remains a largely unexplored and challenging area. In this work, we formulate the versatile safe RL problem and consider two primary requirements: training efficiency and zero-shot adaptation capability. To address them, we introduce the Conditioned Constrained Policy Optimization (CCPO) framework, consisting of two key modules: (1) Versatile Va"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.03718","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/2310.03718/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":"2310.03718","created_at":"2026-07-05T08:13:27.360154+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.03718v2","created_at":"2026-07-05T08:13:27.360154+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.03718","created_at":"2026-07-05T08:13:27.360154+00:00"},{"alias_kind":"pith_short_12","alias_value":"UFTTKZ4GK7L3","created_at":"2026-07-05T08:13:27.360154+00:00"},{"alias_kind":"pith_short_16","alias_value":"UFTTKZ4GK7L3BZGJ","created_at":"2026-07-05T08:13:27.360154+00:00"},{"alias_kind":"pith_short_8","alias_value":"UFTTKZ4G","created_at":"2026-07-05T08:13:27.360154+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/UFTTKZ4GK7L3BZGJMUGAB6RJHT","json":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT.json","graph_json":"https://pith.science/api/pith-number/UFTTKZ4GK7L3BZGJMUGAB6RJHT/graph.json","events_json":"https://pith.science/api/pith-number/UFTTKZ4GK7L3BZGJMUGAB6RJHT/events.json","paper":"https://pith.science/paper/UFTTKZ4G"},"agent_actions":{"view_html":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT","download_json":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT.json","view_paper":"https://pith.science/paper/UFTTKZ4G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.03718&json=true","fetch_graph":"https://pith.science/api/pith-number/UFTTKZ4GK7L3BZGJMUGAB6RJHT/graph.json","fetch_events":"https://pith.science/api/pith-number/UFTTKZ4GK7L3BZGJMUGAB6RJHT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT/action/storage_attestation","attest_author":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT/action/author_attestation","sign_citation":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT/action/citation_signature","submit_replication":"https://pith.science/pith/UFTTKZ4GK7L3BZGJMUGAB6RJHT/action/replication_record"}},"created_at":"2026-07-05T08:13:27.360154+00:00","updated_at":"2026-07-05T08:13:27.360154+00:00"}