{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:D5FPSTRLPXQNR2GLF5UCR44XHP","short_pith_number":"pith:D5FPSTRL","canonical_record":{"source":{"id":"2501.05501","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-09T18:43:05Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"8a459c1f734e93fb52b19146eaf95e0925d8a5caace40104954e7aab5d357454","abstract_canon_sha256":"20484a7c46319f9434fd032d8ca9c631ea2c7d15eb4c72b12cd0fd8f2fdfbea9"},"schema_version":"1.0"},"canonical_sha256":"1f4af94e2b7de0d8e8cb2f6828f3973bec3a95cd4d2a873f759976bd9c0c453b","source":{"kind":"arxiv","id":"2501.05501","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.05501","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"arxiv_version","alias_value":"2501.05501v2","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05501","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"pith_short_12","alias_value":"D5FPSTRLPXQN","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"pith_short_16","alias_value":"D5FPSTRLPXQNR2GL","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"pith_short_8","alias_value":"D5FPSTRL","created_at":"2026-07-05T10:02:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:D5FPSTRLPXQNR2GLF5UCR44XHP","target":"record","payload":{"canonical_record":{"source":{"id":"2501.05501","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-09T18:43:05Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"8a459c1f734e93fb52b19146eaf95e0925d8a5caace40104954e7aab5d357454","abstract_canon_sha256":"20484a7c46319f9434fd032d8ca9c631ea2c7d15eb4c72b12cd0fd8f2fdfbea9"},"schema_version":"1.0"},"canonical_sha256":"1f4af94e2b7de0d8e8cb2f6828f3973bec3a95cd4d2a873f759976bd9c0c453b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:57.679881Z","signature_b64":"bzXfH6P+SxzjCOio9EsxjyE8J8buAXtpT+UcNYKoRODqYxGpKkz2IZUH5icIOIH+Ke9/hpQqBevTRrA5dEmEAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1f4af94e2b7de0d8e8cb2f6828f3973bec3a95cd4d2a873f759976bd9c0c453b","last_reissued_at":"2026-07-05T10:02:57.679325Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:57.679325Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.05501","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:02:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1ov/0z5+TZv7GPAIknLsabIOhE2odqqkskZpYT0Tt9khE6C6/FMD4lMYRCBSr7SZZ94svENqHE4NFXTrqsyDDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T17:06:28.255900Z"},"content_sha256":"64b9cc6861882bd946dc83281142f2734b0af70460a54cf95ce78fd7980a466c","schema_version":"1.0","event_id":"sha256:64b9cc6861882bd946dc83281142f2734b0af70460a54cf95ce78fd7980a466c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:D5FPSTRLPXQNR2GLF5UCR44XHP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Strategy Masking: A Method for Guardrails in Value-based Reinforcement Learning Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Jeremy Kedziora, Jonathan Keane, Sam Keyser","submitted_at":"2025-01-09T18:43:05Z","abstract_excerpt":"The use of reward functions to structure AI learning and decision making is core to the current reinforcement learning paradigm; however, without careful design of reward functions, agents can learn to solve problems in ways that may be considered \"undesirable\" or \"unethical.\" Without thorough understanding of the incentives a reward function creates, it can be difficult to impose principled yet general control mechanisms over its behavior. In this paper, we study methods for constructing guardrails for AI agents that use reward functions to learn decision making. We introduce a novel approach"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05501","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/2501.05501/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":"2026-07-05T10:02:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NgMhe7vwuqRg62vP5F7Sd1dHwBmv2Fm8gmzzpI7YRk5S/Qwp6LS5NYElxI8ETUUQ+0Z+BpuRgVQFdFrCe7sjBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T17:06:28.256437Z"},"content_sha256":"42b1b5a01bffaed72d4bf7b802fd0f59c8af65ad9e877eed1936b35f23203e96","schema_version":"1.0","event_id":"sha256:42b1b5a01bffaed72d4bf7b802fd0f59c8af65ad9e877eed1936b35f23203e96"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/D5FPSTRLPXQNR2GLF5UCR44XHP/bundle.json","state_url":"https://pith.science/pith/D5FPSTRLPXQNR2GLF5UCR44XHP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/D5FPSTRLPXQNR2GLF5UCR44XHP/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":"2026-08-11T17:06:28Z","links":{"resolver":"https://pith.science/pith/D5FPSTRLPXQNR2GLF5UCR44XHP","bundle":"https://pith.science/pith/D5FPSTRLPXQNR2GLF5UCR44XHP/bundle.json","state":"https://pith.science/pith/D5FPSTRLPXQNR2GLF5UCR44XHP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/D5FPSTRLPXQNR2GLF5UCR44XHP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:D5FPSTRLPXQNR2GLF5UCR44XHP","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":"20484a7c46319f9434fd032d8ca9c631ea2c7d15eb4c72b12cd0fd8f2fdfbea9","cross_cats_sorted":["cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-09T18:43:05Z","title_canon_sha256":"8a459c1f734e93fb52b19146eaf95e0925d8a5caace40104954e7aab5d357454"},"schema_version":"1.0","source":{"id":"2501.05501","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.05501","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"arxiv_version","alias_value":"2501.05501v2","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05501","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"pith_short_12","alias_value":"D5FPSTRLPXQN","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"pith_short_16","alias_value":"D5FPSTRLPXQNR2GL","created_at":"2026-07-05T10:02:57Z"},{"alias_kind":"pith_short_8","alias_value":"D5FPSTRL","created_at":"2026-07-05T10:02:57Z"}],"graph_snapshots":[{"event_id":"sha256:42b1b5a01bffaed72d4bf7b802fd0f59c8af65ad9e877eed1936b35f23203e96","target":"graph","created_at":"2026-07-05T10:02:57Z","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/2501.05501/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The use of reward functions to structure AI learning and decision making is core to the current reinforcement learning paradigm; however, without careful design of reward functions, agents can learn to solve problems in ways that may be considered \"undesirable\" or \"unethical.\" Without thorough understanding of the incentives a reward function creates, it can be difficult to impose principled yet general control mechanisms over its behavior. In this paper, we study methods for constructing guardrails for AI agents that use reward functions to learn decision making. We introduce a novel approach","authors_text":"Jeremy Kedziora, Jonathan Keane, Sam Keyser","cross_cats":["cs.LG","cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-09T18:43:05Z","title":"Strategy Masking: A Method for Guardrails in Value-based Reinforcement Learning Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05501","kind":"arxiv","version":2},"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":"sha256:64b9cc6861882bd946dc83281142f2734b0af70460a54cf95ce78fd7980a466c","target":"record","created_at":"2026-07-05T10:02:57Z","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":"20484a7c46319f9434fd032d8ca9c631ea2c7d15eb4c72b12cd0fd8f2fdfbea9","cross_cats_sorted":["cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-09T18:43:05Z","title_canon_sha256":"8a459c1f734e93fb52b19146eaf95e0925d8a5caace40104954e7aab5d357454"},"schema_version":"1.0","source":{"id":"2501.05501","kind":"arxiv","version":2}},"canonical_sha256":"1f4af94e2b7de0d8e8cb2f6828f3973bec3a95cd4d2a873f759976bd9c0c453b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1f4af94e2b7de0d8e8cb2f6828f3973bec3a95cd4d2a873f759976bd9c0c453b","first_computed_at":"2026-07-05T10:02:57.679325Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:02:57.679325Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bzXfH6P+SxzjCOio9EsxjyE8J8buAXtpT+UcNYKoRODqYxGpKkz2IZUH5icIOIH+Ke9/hpQqBevTRrA5dEmEAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:02:57.679881Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.05501","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:64b9cc6861882bd946dc83281142f2734b0af70460a54cf95ce78fd7980a466c","sha256:42b1b5a01bffaed72d4bf7b802fd0f59c8af65ad9e877eed1936b35f23203e96"],"state_sha256":"0448ba102a4587d80bd674e390ead82417652462d539bd0ac26a51380fd29ebd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2Y/QW3VwHyPOgPk0C3DIgBPY+jiwaN8nm4LNCQC0/nnbqxqzKzlVDBZPwycC0aEAD1LY5daecufJT6WMAsokBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T17:06:28.262349Z","bundle_sha256":"e662d38aa12441d5a0f5a6ee4aa12e9cfbbc3529df4141e3eb3bf1e85356307a"}}