{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2IEU4N5BQDITH2APSFKOQZ5NAZ","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"4a32da4e3f64265ff258e6b40a8709afa1f60c1cb3c24e0c2bff422e493b397b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-05T14:29:18Z","title_canon_sha256":"e74c4d411d744b588d67271890ede775482d29c1ad9c815e1c962543b11fedea"},"schema_version":"1.0","source":{"id":"2607.04332","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.04332","created_at":"2026-07-07T02:19:09Z"},{"alias_kind":"arxiv_version","alias_value":"2607.04332v1","created_at":"2026-07-07T02:19:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.04332","created_at":"2026-07-07T02:19:09Z"},{"alias_kind":"pith_short_12","alias_value":"2IEU4N5BQDIT","created_at":"2026-07-07T02:19:09Z"},{"alias_kind":"pith_short_16","alias_value":"2IEU4N5BQDITH2AP","created_at":"2026-07-07T02:19:09Z"},{"alias_kind":"pith_short_8","alias_value":"2IEU4N5B","created_at":"2026-07-07T02:19:09Z"}],"graph_snapshots":[{"event_id":"sha256:ceed08f94d8e61e2c493f378866f595208e9bd9e0f989f388870fc3c8404bef6","target":"graph","created_at":"2026-07-07T02:19:09Z","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/2607.04332/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this paper, we consider the setting where large language models (LLMs) are trained using reinforcement learning (RL) to simultaneously improve reasoning accuracy and verbalize its confidence. Our reward scheme uses two functions for rewarding confidence verbalized by the LLM: one when the LLM is correct and a different one when the LLM is incorrect. With a poorly designed reward scheme, the LLM may be incentivized to answer incorrectly so that it can be confident that its answer is indeed incorrect, a phenomenon that we call confidence reward hacking. We propose the concept of non-hackable ","authors_text":"Chee Heng Tan, Mehul Motani, Wee Sun Lee, Zhuoyi Lin","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-05T14:29:18Z","title":"On the effectiveness of reward functions in reinforcement learning for confidence calibration of large language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.04332","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":"sha256:751173da1e39b4f081ab944c58b1a970a7b6bf7399d0de8469bf6b21385a6070","target":"record","created_at":"2026-07-07T02:19:09Z","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":"4a32da4e3f64265ff258e6b40a8709afa1f60c1cb3c24e0c2bff422e493b397b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-05T14:29:18Z","title_canon_sha256":"e74c4d411d744b588d67271890ede775482d29c1ad9c815e1c962543b11fedea"},"schema_version":"1.0","source":{"id":"2607.04332","kind":"arxiv","version":1}},"canonical_sha256":"d2094e37a180d133e80f9154e867ad0659f29db4caa893524695a983bf9b1a35","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d2094e37a180d133e80f9154e867ad0659f29db4caa893524695a983bf9b1a35","first_computed_at":"2026-07-07T02:19:09.510100Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-07T02:19:09.510100Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cPpbDQ0X+6K4JSZVlbugarkvgnxJu8E0axOcLwnAxbRKcVepZcAAnOEVckKDEKvEmInsKi78/FqSY6rSrMMxCA==","signature_status":"signed_v1","signed_at":"2026-07-07T02:19:09.510762Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.04332","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:751173da1e39b4f081ab944c58b1a970a7b6bf7399d0de8469bf6b21385a6070","sha256:ceed08f94d8e61e2c493f378866f595208e9bd9e0f989f388870fc3c8404bef6","sha256:cf78d1da7cedc909ea967e67218561b9660ca22d525c678b982b9b058c0d4735"],"state_sha256":"d5f579e947025bd90324d591f6ab8ba48d0e504dcb04571f816249b9a488c7cd"}