{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:D44KJPFMJW34VK6UUL43PFJTUW","short_pith_number":"pith:D44KJPFM","canonical_record":{"source":{"id":"2607.13172","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-14T18:22:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6140ce89764a0cab08625998831586d311950a1fcd136ea39ae85bae562e7a12","abstract_canon_sha256":"b1f0749229bf7b83a52b2b9cbf007ff251cb676631d5180d272d21cb286769ba"},"schema_version":"1.0"},"canonical_sha256":"1f38a4bcac4db7caabd4a2f9b79533a5862a9e67d3b24637a93742449ad7ceaf","source":{"kind":"arxiv","id":"2607.13172","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.13172","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"arxiv_version","alias_value":"2607.13172v1","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.13172","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"pith_short_12","alias_value":"D44KJPFMJW34","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"pith_short_16","alias_value":"D44KJPFMJW34VK6U","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"pith_short_8","alias_value":"D44KJPFM","created_at":"2026-07-16T00:22:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:D44KJPFMJW34VK6UUL43PFJTUW","target":"record","payload":{"canonical_record":{"source":{"id":"2607.13172","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-14T18:22:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6140ce89764a0cab08625998831586d311950a1fcd136ea39ae85bae562e7a12","abstract_canon_sha256":"b1f0749229bf7b83a52b2b9cbf007ff251cb676631d5180d272d21cb286769ba"},"schema_version":"1.0"},"canonical_sha256":"1f38a4bcac4db7caabd4a2f9b79533a5862a9e67d3b24637a93742449ad7ceaf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-16T00:22:01.080717Z","signature_b64":"Rdp8vltfjWZ9bbPKUzGNs2CaCzdA/bKdwArD7i0bzRnYVl2FtFFwWFlD+hWMu7F1/4HGYuhEq+YPMI8QHboEBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1f38a4bcac4db7caabd4a2f9b79533a5862a9e67d3b24637a93742449ad7ceaf","last_reissued_at":"2026-07-16T00:22:01.079905Z","signature_status":"signed_v1","first_computed_at":"2026-07-16T00:22:01.079905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.13172","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":"2026-07-16T00:22:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HFSJ/hDC97GBO80l/g16bNneEsjbdWR1/cO1hhizDuuZIxF4JWE46g39ZSpt35a5ZCU7XTK4tSjVlAzHtJXbAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T12:36:00.418495Z"},"content_sha256":"eabd2cab0e487a43a0d5fab2bf7f2737a821549e6758e6eb0a0ffe2b01601f2c","schema_version":"1.0","event_id":"sha256:eabd2cab0e487a43a0d5fab2bf7f2737a821549e6758e6eb0a0ffe2b01601f2c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:D44KJPFMJW34VK6UUL43PFJTUW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning Safe Agent Behaviour from Human Preferences and Justifications via World Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Christopher T. Freeman, Ilias Kazantzidis, Timothy J. Norman, Yali Du","submitted_at":"2026-07-14T18:22:14Z","abstract_excerpt":"We address the problem of safely training an agent policy and deploying a good and safe policy, in settings where the environment dynamics are unknown and no suitable reward function is available. In the context of safety-critical environments, we consider traditional reinforcement learning impractical and resort to the resource of human input. We introduce DROPJ, a human-centred method for both safe training and deployment. We first learn a world model (a learned simulator) from a dataset of prior real-world trajectories. A human then plays the game in this learned simulator to extract severa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.13172","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/2607.13172/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":"2026-07-16T00:22:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D0lOlQ4xq5qVYYStMfqHlv7BWs52w747iKgMNY8QKbxBMcWkbp15e82CldzEEVpfsfvGpBQRNzfOrjosGN+CAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T12:36:00.419006Z"},"content_sha256":"0118446cce556e5e73514c25ee6052d553982a1226ee22c6fbb608c4ec94a1ba","schema_version":"1.0","event_id":"sha256:0118446cce556e5e73514c25ee6052d553982a1226ee22c6fbb608c4ec94a1ba"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/D44KJPFMJW34VK6UUL43PFJTUW/bundle.json","state_url":"https://pith.science/pith/D44KJPFMJW34VK6UUL43PFJTUW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/D44KJPFMJW34VK6UUL43PFJTUW/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":"2026-08-23T12:36:00Z","links":{"resolver":"https://pith.science/pith/D44KJPFMJW34VK6UUL43PFJTUW","bundle":"https://pith.science/pith/D44KJPFMJW34VK6UUL43PFJTUW/bundle.json","state":"https://pith.science/pith/D44KJPFMJW34VK6UUL43PFJTUW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/D44KJPFMJW34VK6UUL43PFJTUW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:D44KJPFMJW34VK6UUL43PFJTUW","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":"b1f0749229bf7b83a52b2b9cbf007ff251cb676631d5180d272d21cb286769ba","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-14T18:22:14Z","title_canon_sha256":"6140ce89764a0cab08625998831586d311950a1fcd136ea39ae85bae562e7a12"},"schema_version":"1.0","source":{"id":"2607.13172","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.13172","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"arxiv_version","alias_value":"2607.13172v1","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.13172","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"pith_short_12","alias_value":"D44KJPFMJW34","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"pith_short_16","alias_value":"D44KJPFMJW34VK6U","created_at":"2026-07-16T00:22:01Z"},{"alias_kind":"pith_short_8","alias_value":"D44KJPFM","created_at":"2026-07-16T00:22:01Z"}],"graph_snapshots":[{"event_id":"sha256:0118446cce556e5e73514c25ee6052d553982a1226ee22c6fbb608c4ec94a1ba","target":"graph","created_at":"2026-07-16T00:22:01Z","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/2607.13172/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We address the problem of safely training an agent policy and deploying a good and safe policy, in settings where the environment dynamics are unknown and no suitable reward function is available. In the context of safety-critical environments, we consider traditional reinforcement learning impractical and resort to the resource of human input. We introduce DROPJ, a human-centred method for both safe training and deployment. We first learn a world model (a learned simulator) from a dataset of prior real-world trajectories. A human then plays the game in this learned simulator to extract severa","authors_text":"Christopher T. Freeman, Ilias Kazantzidis, Timothy J. Norman, Yali Du","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-14T18:22:14Z","title":"Learning Safe Agent Behaviour from Human Preferences and Justifications via World Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.13172","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:eabd2cab0e487a43a0d5fab2bf7f2737a821549e6758e6eb0a0ffe2b01601f2c","target":"record","created_at":"2026-07-16T00:22:01Z","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":"b1f0749229bf7b83a52b2b9cbf007ff251cb676631d5180d272d21cb286769ba","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-14T18:22:14Z","title_canon_sha256":"6140ce89764a0cab08625998831586d311950a1fcd136ea39ae85bae562e7a12"},"schema_version":"1.0","source":{"id":"2607.13172","kind":"arxiv","version":1}},"canonical_sha256":"1f38a4bcac4db7caabd4a2f9b79533a5862a9e67d3b24637a93742449ad7ceaf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1f38a4bcac4db7caabd4a2f9b79533a5862a9e67d3b24637a93742449ad7ceaf","first_computed_at":"2026-07-16T00:22:01.079905Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-16T00:22:01.079905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Rdp8vltfjWZ9bbPKUzGNs2CaCzdA/bKdwArD7i0bzRnYVl2FtFFwWFlD+hWMu7F1/4HGYuhEq+YPMI8QHboEBQ==","signature_status":"signed_v1","signed_at":"2026-07-16T00:22:01.080717Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.13172","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eabd2cab0e487a43a0d5fab2bf7f2737a821549e6758e6eb0a0ffe2b01601f2c","sha256:0118446cce556e5e73514c25ee6052d553982a1226ee22c6fbb608c4ec94a1ba"],"state_sha256":"8c58e615071dff5eb5cd3a554ff814460441a2532740c1bfb0ffa8355d6a8327"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oVM7JtflwTbPC3QD1SylP2DcyeoK2cJpC31dm/s01/9gWSGeLb9+i04g2Newp3OuEQZ+Ni/kXEFyr6GhPm6RBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T12:36:00.424750Z","bundle_sha256":"b03b2f46508d50e98f81f494b2c1a433af6ea4720a13009906221baf6e1d526e"}}