{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RFK3W6J7SKYF426666JQFV7MMB","short_pith_number":"pith:RFK3W6J7","canonical_record":{"source":{"id":"2502.20341","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T18:10:33Z","cross_cats_sorted":[],"title_canon_sha256":"9f106f05ff8c15041887f2a2987767b6a9e25ab0728e7779258a10599914d2e8","abstract_canon_sha256":"e208ce1bcb62d2db96aa87975d777585486b22608e76ad4a86cf9257ae1600d9"},"schema_version":"1.0"},"canonical_sha256":"8955bb793f92b05e6bdef79302d7ec607ff2e3c379550aefab8b7b689a96c02e","source":{"kind":"arxiv","id":"2502.20341","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20341","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20341v1","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20341","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"pith_short_12","alias_value":"RFK3W6J7SKYF","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"pith_short_16","alias_value":"RFK3W6J7SKYF4266","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"pith_short_8","alias_value":"RFK3W6J7","created_at":"2026-07-05T10:21:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RFK3W6J7SKYF426666JQFV7MMB","target":"record","payload":{"canonical_record":{"source":{"id":"2502.20341","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T18:10:33Z","cross_cats_sorted":[],"title_canon_sha256":"9f106f05ff8c15041887f2a2987767b6a9e25ab0728e7779258a10599914d2e8","abstract_canon_sha256":"e208ce1bcb62d2db96aa87975d777585486b22608e76ad4a86cf9257ae1600d9"},"schema_version":"1.0"},"canonical_sha256":"8955bb793f92b05e6bdef79302d7ec607ff2e3c379550aefab8b7b689a96c02e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:13.242939Z","signature_b64":"V1FxoLb0RvYZZTVOaCZVNr+OVKslgOmGrVGZo9TEGiZdQG97r+vZM5sWh/M1MuimYH70Yky7j0Q4CoLNpDsRDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8955bb793f92b05e6bdef79302d7ec607ff2e3c379550aefab8b7b689a96c02e","last_reissued_at":"2026-07-05T10:21:13.240340Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:13.240340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.20341","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:21:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g6N0t0mBG6jdP9l/ZeRqL8puU2Lf4RxDxPR5wABGwr2d07sE0bntQrh/Gj0Kr/Qof065FbWtwz3cZfOuZyoOBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T07:32:00.657860Z"},"content_sha256":"005ff569a5a68f56d02f0cd73cb6d6e897c32f7c8035d1d25689b54b1bed12e7","schema_version":"1.0","event_id":"sha256:005ff569a5a68f56d02f0cd73cb6d6e897c32f7c8035d1d25689b54b1bed12e7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RFK3W6J7SKYF426666JQFV7MMB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Safety Representations for Safer Policy Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Annie Chen, Charlie Gauthier, Kaustubh Mani, Liam Paull, Samer Nashed, Vincent Mai","submitted_at":"2025-02-27T18:10:33Z","abstract_excerpt":"Reinforcement learning algorithms typically necessitate extensive exploration of the state space to find optimal policies. However, in safety-critical applications, the risks associated with such exploration can lead to catastrophic consequences. Existing safe exploration methods attempt to mitigate this by imposing constraints, which often result in overly conservative behaviours and inefficient learning. Heavy penalties for early constraint violations can trap agents in local optima, deterring exploration of risky yet high-reward regions of the state space. To address this, we introduce a me"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20341","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/2502.20341/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:21:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wkUQRoUS/XUqwUf1GX46rP/M/0XxIkCgsXgAyevX+yO6Ti4+6GSsyO/v9DLFEnEh4w6+eLpGlFegRG2YGB6HDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T07:32:00.658468Z"},"content_sha256":"34b9475737d2b0f3989c66d357e8f6f424eef67201d8424ef92a9be280fdd898","schema_version":"1.0","event_id":"sha256:34b9475737d2b0f3989c66d357e8f6f424eef67201d8424ef92a9be280fdd898"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RFK3W6J7SKYF426666JQFV7MMB/bundle.json","state_url":"https://pith.science/pith/RFK3W6J7SKYF426666JQFV7MMB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RFK3W6J7SKYF426666JQFV7MMB/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T07:32:00Z","links":{"resolver":"https://pith.science/pith/RFK3W6J7SKYF426666JQFV7MMB","bundle":"https://pith.science/pith/RFK3W6J7SKYF426666JQFV7MMB/bundle.json","state":"https://pith.science/pith/RFK3W6J7SKYF426666JQFV7MMB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RFK3W6J7SKYF426666JQFV7MMB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RFK3W6J7SKYF426666JQFV7MMB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e208ce1bcb62d2db96aa87975d777585486b22608e76ad4a86cf9257ae1600d9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T18:10:33Z","title_canon_sha256":"9f106f05ff8c15041887f2a2987767b6a9e25ab0728e7779258a10599914d2e8"},"schema_version":"1.0","source":{"id":"2502.20341","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20341","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20341v1","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20341","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"pith_short_12","alias_value":"RFK3W6J7SKYF","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"pith_short_16","alias_value":"RFK3W6J7SKYF4266","created_at":"2026-07-05T10:21:13Z"},{"alias_kind":"pith_short_8","alias_value":"RFK3W6J7","created_at":"2026-07-05T10:21:13Z"}],"graph_snapshots":[{"event_id":"sha256:34b9475737d2b0f3989c66d357e8f6f424eef67201d8424ef92a9be280fdd898","target":"graph","created_at":"2026-07-05T10:21:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.20341/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning algorithms typically necessitate extensive exploration of the state space to find optimal policies. However, in safety-critical applications, the risks associated with such exploration can lead to catastrophic consequences. Existing safe exploration methods attempt to mitigate this by imposing constraints, which often result in overly conservative behaviours and inefficient learning. Heavy penalties for early constraint violations can trap agents in local optima, deterring exploration of risky yet high-reward regions of the state space. To address this, we introduce a me","authors_text":"Annie Chen, Charlie Gauthier, Kaustubh Mani, Liam Paull, Samer Nashed, Vincent Mai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T18:10:33Z","title":"Safety Representations for Safer Policy Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20341","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:005ff569a5a68f56d02f0cd73cb6d6e897c32f7c8035d1d25689b54b1bed12e7","target":"record","created_at":"2026-07-05T10:21:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e208ce1bcb62d2db96aa87975d777585486b22608e76ad4a86cf9257ae1600d9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T18:10:33Z","title_canon_sha256":"9f106f05ff8c15041887f2a2987767b6a9e25ab0728e7779258a10599914d2e8"},"schema_version":"1.0","source":{"id":"2502.20341","kind":"arxiv","version":1}},"canonical_sha256":"8955bb793f92b05e6bdef79302d7ec607ff2e3c379550aefab8b7b689a96c02e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8955bb793f92b05e6bdef79302d7ec607ff2e3c379550aefab8b7b689a96c02e","first_computed_at":"2026-07-05T10:21:13.240340Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:13.240340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"V1FxoLb0RvYZZTVOaCZVNr+OVKslgOmGrVGZo9TEGiZdQG97r+vZM5sWh/M1MuimYH70Yky7j0Q4CoLNpDsRDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:13.242939Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.20341","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:005ff569a5a68f56d02f0cd73cb6d6e897c32f7c8035d1d25689b54b1bed12e7","sha256:34b9475737d2b0f3989c66d357e8f6f424eef67201d8424ef92a9be280fdd898"],"state_sha256":"a5cfced9796b5ed29b72770ed04cd698537500cb18bf90cc8ed3885432a52df1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GBV1ZSDD4QcWmEA4/qlxRRcumyIV30k0S4THXHXQlitYMEmbQHOFlHdHmeXUAQ/j70Qp0DPAN4LU4ODrKoaJDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T07:32:00.660990Z","bundle_sha256":"000aa0e1bac3947aedffaa8b859a2364772f3b1a3695a52a87690ff2f24e1f34"}}