{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JADZI3HULM4JKPYGCRHZZ5P7ED","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":"4ab1f9c4f9d1e598adf8a23394c822da821917faf0c129c908a61a687adb7bee","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-02-06T00:11:37Z","title_canon_sha256":"edf5b7c296b5c77d4b43d951b90db0ee5eb940addd666a3a0ce0f576e5d40f75"},"schema_version":"1.0","source":{"id":"2603.03291","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.03291","created_at":"2026-06-02T02:04:15Z"},{"alias_kind":"arxiv_version","alias_value":"2603.03291v2","created_at":"2026-06-02T02:04:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.03291","created_at":"2026-06-02T02:04:15Z"},{"alias_kind":"pith_short_12","alias_value":"JADZI3HULM4J","created_at":"2026-06-02T02:04:15Z"},{"alias_kind":"pith_short_16","alias_value":"JADZI3HULM4JKPYG","created_at":"2026-06-02T02:04:15Z"},{"alias_kind":"pith_short_8","alias_value":"JADZI3HU","created_at":"2026-06-02T02:04:15Z"}],"graph_snapshots":[{"event_id":"sha256:0874a6f857a6688f400d94a4762ac1522123309cc6f888fb5832ae0856fbd810","target":"graph","created_at":"2026-06-02T02:04:15Z","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/2603.03291/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward Models (RMs) are crucial for online alignment of language models (LMs) with human preferences. However, RM-based preference-tuning is vulnerable to reward hacking, whereby LM policies learn undesirable behaviors from flawed RMs. By systematically measuring biases in five high-quality RMs, including the state-of-the-art, we find that issues persist despite prior work with respect to length, sycophancy, and overconfidence. We also discover new issues related to bias toward model-specific ``styles'' and answer-order. We categorize RM failures as tractable or resistant to linear interventio","authors_text":"Daniel Fein, Max Lamparth, Mykel J. Kochenderfer, Nick Haber, Violet Xiang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-02-06T00:11:37Z","title":"One Bias After Another: Mechanistic Reward Shaping and Persistent Biases in Language Reward Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.03291","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:b09dafb69e524cd1f538417970ad958d6a9e539c62269ee01286ceb42a66d68f","target":"record","created_at":"2026-06-02T02:04:15Z","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":"4ab1f9c4f9d1e598adf8a23394c822da821917faf0c129c908a61a687adb7bee","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-02-06T00:11:37Z","title_canon_sha256":"edf5b7c296b5c77d4b43d951b90db0ee5eb940addd666a3a0ce0f576e5d40f75"},"schema_version":"1.0","source":{"id":"2603.03291","kind":"arxiv","version":2}},"canonical_sha256":"4807946cf45b38953f06144f9cf5ff20eaf0381146e3429f92d55a356d40c007","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4807946cf45b38953f06144f9cf5ff20eaf0381146e3429f92d55a356d40c007","first_computed_at":"2026-06-02T02:04:15.551574Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:15.551574Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QXKzmZKkGaTNNWnQe+zwkcv5Bqd/Nlf7di9865ja/RsKuixptjIRN12XNVDHZMftQUv80n6mSXs1PLC7w6W+DQ==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:15.552086Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.03291","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b09dafb69e524cd1f538417970ad958d6a9e539c62269ee01286ceb42a66d68f","sha256:0874a6f857a6688f400d94a4762ac1522123309cc6f888fb5832ae0856fbd810"],"state_sha256":"80c39f4812431b10f52a87e95ceaea4173fd31eeceed61274cc1cf5d8378361d"}