{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NRKCGJPWCQZWG2RP5Y2XTRWFAX","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":"c2f0fcaed077517bb23409556f821b9595f417987641f4028037494b03ff1ae7","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T16:32:54Z","title_canon_sha256":"383fe7649f3826be25a6d2a2ab7f7fc7de420b4005a06d714b22a784b6ee908c"},"schema_version":"1.0","source":{"id":"2606.09711","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09711","created_at":"2026-06-09T02:09:05Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09711v1","created_at":"2026-06-09T02:09:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09711","created_at":"2026-06-09T02:09:05Z"},{"alias_kind":"pith_short_12","alias_value":"NRKCGJPWCQZW","created_at":"2026-06-09T02:09:05Z"},{"alias_kind":"pith_short_16","alias_value":"NRKCGJPWCQZWG2RP","created_at":"2026-06-09T02:09:05Z"},{"alias_kind":"pith_short_8","alias_value":"NRKCGJPW","created_at":"2026-06-09T02:09:05Z"}],"graph_snapshots":[{"event_id":"sha256:8dc84fa6c6f0afe041a87f1b8f4327c8b4bb0aaa168d39d45d1343e09a907333","target":"graph","created_at":"2026-06-09T02:09:05Z","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/2606.09711/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward hacking is usually studied after it becomes visible, once a model earns high proxy reward while failing the intended task. We instead study what proxy RL teaches before that failure appears. We introduce Proxy Reward Internalization and Mechanistic Exploitation (PRIME), a learned capability to assess task correctness, predict proxy acceptance, and reason about exploitable proxy--gold gaps. In coding RL environments with exploitable pytest rewards, we measure PRIME through chain-of-thought monitoring, direct probes, and activation-level concept vectors. We find that PRIME emerges in a st","authors_text":"Lifu Huang, Ming Jin, Mohammad Beigi","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T16:32:54Z","title":"Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09711","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:ffdec11a5984812e2edf85b9a5fcbc44c3c568741ef6541bc427ea540ffc6925","target":"record","created_at":"2026-06-09T02:09:05Z","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":"c2f0fcaed077517bb23409556f821b9595f417987641f4028037494b03ff1ae7","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T16:32:54Z","title_canon_sha256":"383fe7649f3826be25a6d2a2ab7f7fc7de420b4005a06d714b22a784b6ee908c"},"schema_version":"1.0","source":{"id":"2606.09711","kind":"arxiv","version":1}},"canonical_sha256":"6c542325f61433636a2fee3579c6c505c1b226439d6f2e9557d3fc22dd46c16b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c542325f61433636a2fee3579c6c505c1b226439d6f2e9557d3fc22dd46c16b","first_computed_at":"2026-06-09T02:09:05.246361Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:09:05.246361Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"F2xh9pFU8hDLhkFrbcc3CUvK5ojjgK3xRu74mMW10WPu2VEZceu87SR8WcnZ0gbn2PFgiAzhybxpq/LvjwK9AQ==","signature_status":"signed_v1","signed_at":"2026-06-09T02:09:05.247252Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.09711","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ffdec11a5984812e2edf85b9a5fcbc44c3c568741ef6541bc427ea540ffc6925","sha256:8dc84fa6c6f0afe041a87f1b8f4327c8b4bb0aaa168d39d45d1343e09a907333"],"state_sha256":"8ee8ab6beadbad12ddf874feeb7c0d99098a4f0c2e1e5a0413f436c3e8c1cb5f"}