{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KOTFVC4KHSF3KMDSLNGP4HO5AB","short_pith_number":"pith:KOTFVC4K","canonical_record":{"source":{"id":"2505.02709","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-05T15:06:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"14ddcd851347fcfc9329e2fe6fee8b88e92c222ed3d01e21199335507d0db1bc","abstract_canon_sha256":"51fbba4c9550f962be3fc7d9679d0d0e27d197bb2e23a5f9a5f03683fd70306b"},"schema_version":"1.0"},"canonical_sha256":"53a65a8b8a3c8bb530725b4cfe1ddd00446dc3ddb0b08b7ff1e728f28e4e7fae","source":{"kind":"arxiv","id":"2505.02709","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.02709","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"arxiv_version","alias_value":"2505.02709v1","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.02709","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"pith_short_12","alias_value":"KOTFVC4KHSF3","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"pith_short_16","alias_value":"KOTFVC4KHSF3KMDS","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"pith_short_8","alias_value":"KOTFVC4K","created_at":"2026-07-05T10:58:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KOTFVC4KHSF3KMDSLNGP4HO5AB","target":"record","payload":{"canonical_record":{"source":{"id":"2505.02709","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-05T15:06:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"14ddcd851347fcfc9329e2fe6fee8b88e92c222ed3d01e21199335507d0db1bc","abstract_canon_sha256":"51fbba4c9550f962be3fc7d9679d0d0e27d197bb2e23a5f9a5f03683fd70306b"},"schema_version":"1.0"},"canonical_sha256":"53a65a8b8a3c8bb530725b4cfe1ddd00446dc3ddb0b08b7ff1e728f28e4e7fae","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:41.658808Z","signature_b64":"3/H+mtF0SBqyeAo2PASIlSPDfrGgy2smL8VgPzlUoc7akvj6yKiamM3316NUI7huFMvPyb++X7iIfOX4n9FiCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53a65a8b8a3c8bb530725b4cfe1ddd00446dc3ddb0b08b7ff1e728f28e4e7fae","last_reissued_at":"2026-07-05T10:58:41.658297Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:41.658297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.02709","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":"2026-07-05T10:58:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Gqpi7S0h/e2EmUsXpLcOys7Gq6Am5IMDFMCOm9xox/LrOUEDqnj14iWNsihjxaH/5kAyE0IcfY1ciIAr9OVZAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T12:38:38.472060Z"},"content_sha256":"c4386fbc95bf0285327f84ac55adffeb4e721b31320aae14a04d0aef1ed5c537","schema_version":"1.0","event_id":"sha256:c4386fbc95bf0285327f84ac55adffeb4e721b31320aae14a04d0aef1ed5c537"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KOTFVC4KHSF3KMDSLNGP4HO5AB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Technical Report: Evaluating Goal Drift in Language Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Elizabeth Donoway, Henning Bartsch, Marius Hobbhahn, Rauno Arike","submitted_at":"2025-05-05T15:06:09Z","abstract_excerpt":"As language models (LMs) are increasingly deployed as autonomous agents, their robust adherence to human-assigned objectives becomes crucial for safe operation. When these agents operate independently for extended periods without human oversight, even initially well-specified goals may gradually shift. Detecting and measuring goal drift - an agent's tendency to deviate from its original objective over time - presents significant challenges, as goals can shift gradually, causing only subtle behavioral changes. This paper proposes a novel approach to analyzing goal drift in LM agents. In our exp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.02709","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/2505.02709/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":"2026-07-05T10:58:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lM5Ix0HzU/ZyPD/RJxRMrGU8TA1qZxZvkERit332cLtQjNOq949wRLOdempvlvk6kps0zx9JH8WOrKh5cjYVDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T12:38:38.472621Z"},"content_sha256":"dbfdcb2b57d1d7ea08b1aa67df39c95e3fe1263f684e1c8b07dcda59d6b04cf5","schema_version":"1.0","event_id":"sha256:dbfdcb2b57d1d7ea08b1aa67df39c95e3fe1263f684e1c8b07dcda59d6b04cf5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB/bundle.json","state_url":"https://pith.science/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB/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":"2026-08-09T12:38:38Z","links":{"resolver":"https://pith.science/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB","bundle":"https://pith.science/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB/bundle.json","state":"https://pith.science/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KOTFVC4KHSF3KMDSLNGP4HO5AB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KOTFVC4KHSF3KMDSLNGP4HO5AB","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":"51fbba4c9550f962be3fc7d9679d0d0e27d197bb2e23a5f9a5f03683fd70306b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-05T15:06:09Z","title_canon_sha256":"14ddcd851347fcfc9329e2fe6fee8b88e92c222ed3d01e21199335507d0db1bc"},"schema_version":"1.0","source":{"id":"2505.02709","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.02709","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"arxiv_version","alias_value":"2505.02709v1","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.02709","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"pith_short_12","alias_value":"KOTFVC4KHSF3","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"pith_short_16","alias_value":"KOTFVC4KHSF3KMDS","created_at":"2026-07-05T10:58:41Z"},{"alias_kind":"pith_short_8","alias_value":"KOTFVC4K","created_at":"2026-07-05T10:58:41Z"}],"graph_snapshots":[{"event_id":"sha256:dbfdcb2b57d1d7ea08b1aa67df39c95e3fe1263f684e1c8b07dcda59d6b04cf5","target":"graph","created_at":"2026-07-05T10:58:41Z","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/2505.02709/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As language models (LMs) are increasingly deployed as autonomous agents, their robust adherence to human-assigned objectives becomes crucial for safe operation. When these agents operate independently for extended periods without human oversight, even initially well-specified goals may gradually shift. Detecting and measuring goal drift - an agent's tendency to deviate from its original objective over time - presents significant challenges, as goals can shift gradually, causing only subtle behavioral changes. This paper proposes a novel approach to analyzing goal drift in LM agents. In our exp","authors_text":"Elizabeth Donoway, Henning Bartsch, Marius Hobbhahn, Rauno Arike","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-05T15:06:09Z","title":"Technical Report: Evaluating Goal Drift in Language Model Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.02709","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:c4386fbc95bf0285327f84ac55adffeb4e721b31320aae14a04d0aef1ed5c537","target":"record","created_at":"2026-07-05T10:58:41Z","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":"51fbba4c9550f962be3fc7d9679d0d0e27d197bb2e23a5f9a5f03683fd70306b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-05T15:06:09Z","title_canon_sha256":"14ddcd851347fcfc9329e2fe6fee8b88e92c222ed3d01e21199335507d0db1bc"},"schema_version":"1.0","source":{"id":"2505.02709","kind":"arxiv","version":1}},"canonical_sha256":"53a65a8b8a3c8bb530725b4cfe1ddd00446dc3ddb0b08b7ff1e728f28e4e7fae","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53a65a8b8a3c8bb530725b4cfe1ddd00446dc3ddb0b08b7ff1e728f28e4e7fae","first_computed_at":"2026-07-05T10:58:41.658297Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:58:41.658297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3/H+mtF0SBqyeAo2PASIlSPDfrGgy2smL8VgPzlUoc7akvj6yKiamM3316NUI7huFMvPyb++X7iIfOX4n9FiCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:58:41.658808Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.02709","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c4386fbc95bf0285327f84ac55adffeb4e721b31320aae14a04d0aef1ed5c537","sha256:dbfdcb2b57d1d7ea08b1aa67df39c95e3fe1263f684e1c8b07dcda59d6b04cf5"],"state_sha256":"8b70c26fd0f060e1a35feaea959c4e5bce37327b600c7f93dd05ff43509d7c75"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7yYlgRAs3+n+/p9g2Vjsbu+7MC+nS/lXQ0us64SSb6S5k+SNDY1uCSAnXyIcbAA9chQaQm1fxSs2lWROQoxtAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T12:38:38.478468Z","bundle_sha256":"f8312efeaaa7f418a5120a4292c179f0421ba461278c9700d98e2daba67cfea5"}}