{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4P5XPPGT34NNX4JBXIBDBAJTEE","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":"addf39ec5673ddb5ee455b4a182d6c94da62873a50ec90ca99c7c1509b12055c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T13:52:55Z","title_canon_sha256":"6a10ee1a862147779c215bc17d93c904d2a67102d9bedd5baa08742cf9dd9426"},"schema_version":"1.0","source":{"id":"2506.01692","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01692","created_at":"2026-07-05T11:14:17Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01692v1","created_at":"2026-07-05T11:14:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01692","created_at":"2026-07-05T11:14:17Z"},{"alias_kind":"pith_short_12","alias_value":"4P5XPPGT34NN","created_at":"2026-07-05T11:14:17Z"},{"alias_kind":"pith_short_16","alias_value":"4P5XPPGT34NNX4JB","created_at":"2026-07-05T11:14:17Z"},{"alias_kind":"pith_short_8","alias_value":"4P5XPPGT","created_at":"2026-07-05T11:14:17Z"}],"graph_snapshots":[{"event_id":"sha256:a4eab5a3c7d7a3950f0a79b6cfe18300d0e8325ed71ca04144008d16ed814213","target":"graph","created_at":"2026-07-05T11:14:17Z","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/2506.01692/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human preferences in RLHF are typically modeled as a function of the human's reward function or corresponding optimal state-action values. In this work, we propose that human beliefs about the capabilities of the agent being trained also play a key role in preference generation. We examine two questions related to this hypothesis, one descriptive and one normative, respectively: Do human labelers' beliefs about agent capabilities affect the preferences that they provide? And what is the ideal set of beliefs about an agent -- and resulting preferences -- for humans to have? We propose a new pre","authors_text":"Frank Chiu, Scott Niekum, Shripad Deshmukh, Sylee Dandekar, W. Bradley Knox","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T13:52:55Z","title":"A Descriptive and Normative Theory of Human Beliefs in RLHF"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01692","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":"sha256:3fd79a6b5465b8292f252a4e597d23072ac2a1b432f44f0570f030c4f5a2624f","target":"record","created_at":"2026-07-05T11:14:17Z","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":"addf39ec5673ddb5ee455b4a182d6c94da62873a50ec90ca99c7c1509b12055c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T13:52:55Z","title_canon_sha256":"6a10ee1a862147779c215bc17d93c904d2a67102d9bedd5baa08742cf9dd9426"},"schema_version":"1.0","source":{"id":"2506.01692","kind":"arxiv","version":1}},"canonical_sha256":"e3fb77bcd3df1adbf121ba023081332135e43686821e54f8e4b9c8adc4485bd2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3fb77bcd3df1adbf121ba023081332135e43686821e54f8e4b9c8adc4485bd2","first_computed_at":"2026-07-05T11:14:17.970897Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:17.970897Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hgoPxfNaGcsvHEKURfZc5YeM52j9ZUgDPSfPKMkkEjksmJIDsaMevqSXF9V+Uid1+VaX8xttWTEXlgTnW/NUCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:17.971374Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01692","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3fd79a6b5465b8292f252a4e597d23072ac2a1b432f44f0570f030c4f5a2624f","sha256:a4eab5a3c7d7a3950f0a79b6cfe18300d0e8325ed71ca04144008d16ed814213"],"state_sha256":"9e72620af941f83ce92bb9f143faaeaf0651bc504ca2e276ade883511506c07d"}