{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:IZQGIGIVLYML5RMO4RRBLJT6U5","short_pith_number":"pith:IZQGIGIV","canonical_record":{"source":{"id":"2311.14743","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-21T18:41:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"57f77f16e7606d73359dd7bc0c8ac6eac1b8df0d6320e99f2d306c700e3aaf5b","abstract_canon_sha256":"6fc52ff3de55ebebb848fa1e663a8942a820a2e1719c9885a070501e8495a3cc"},"schema_version":"1.0"},"canonical_sha256":"46606419155e18bec58ee46215a67ea75503c1c1810d26751a1e176813b6908a","source":{"kind":"arxiv","id":"2311.14743","version":7},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.14743","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"arxiv_version","alias_value":"2311.14743v7","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.14743","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"pith_short_12","alias_value":"IZQGIGIVLYML","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"pith_short_16","alias_value":"IZQGIGIVLYML5RMO","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"pith_short_8","alias_value":"IZQGIGIV","created_at":"2026-07-05T07:36:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:IZQGIGIVLYML5RMO4RRBLJT6U5","target":"record","payload":{"canonical_record":{"source":{"id":"2311.14743","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-21T18:41:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"57f77f16e7606d73359dd7bc0c8ac6eac1b8df0d6320e99f2d306c700e3aaf5b","abstract_canon_sha256":"6fc52ff3de55ebebb848fa1e663a8942a820a2e1719c9885a070501e8495a3cc"},"schema_version":"1.0"},"canonical_sha256":"46606419155e18bec58ee46215a67ea75503c1c1810d26751a1e176813b6908a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:36:56.449850Z","signature_b64":"UYZ0XmVaxaPwbBd+wsyJ0iGP0OTXPLr9o4k9u6ZJwUtTQkxRKKPE53sLcXDoI2rU8gb//E8357hFJMduaBPUDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46606419155e18bec58ee46215a67ea75503c1c1810d26751a1e176813b6908a","last_reissued_at":"2026-07-05T07:36:56.449386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:36:56.449386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.14743","source_version":7,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:36:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9mrEQjAZpwKcKRD4HgI+VI5nhCbD6rS+sgp1kP7+RS2qhQil+og3hv7L+ZgZ5UEMPwrjLWX5JrpOAtFjB7FaCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T09:50:04.110889Z"},"content_sha256":"abc9c23df57687555cbae3b9a64689ab23f6ac64be7115f280cb0aaa92a76f5c","schema_version":"1.0","event_id":"sha256:abc9c23df57687555cbae3b9a64689ab23f6ac64be7115f280cb0aaa92a76f5c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:IZQGIGIVLYML5RMO4RRBLJT6U5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Baseline Analysis of Reward Models' Ability To Accurately Analyze Foundation Models Under Distribution Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Anthony Chen, Benjamin Pikus, Sean Hendryx, Will LeVine","submitted_at":"2023-11-21T18:41:26Z","abstract_excerpt":"Foundation models, specifically Large Language Models (LLMs), have lately gained wide-spread attention and adoption. Reinforcement Learning with Human Feedback (RLHF) involves training a reward model to capture desired behaviors, which is then used to align LLM's. These reward models are additionally used at inference-time to estimate LLM responses' adherence to those desired behaviors. However, there is little work measuring how robust these reward models are to distribution shifts. In this work, we evaluate how reward model performance - measured via accuracy and calibration (i.e. alignment "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.14743","kind":"arxiv","version":7},"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/2311.14743/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":"2026-07-05T07:36:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PIzPxUWX3K1m8zLkM/5RZ7MSvqZVQmWf5TnN2YMn33gGuACCA3qkwYej2BAJiG3tl7TADQ5KS9E8VBPn09yiDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T09:50:04.111996Z"},"content_sha256":"7dfec937026c49e10389fe71e82d58a2e228ba84ca27f7874588e4f5a33b5a7a","schema_version":"1.0","event_id":"sha256:7dfec937026c49e10389fe71e82d58a2e228ba84ca27f7874588e4f5a33b5a7a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IZQGIGIVLYML5RMO4RRBLJT6U5/bundle.json","state_url":"https://pith.science/pith/IZQGIGIVLYML5RMO4RRBLJT6U5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IZQGIGIVLYML5RMO4RRBLJT6U5/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":"2026-08-13T09:50:04Z","links":{"resolver":"https://pith.science/pith/IZQGIGIVLYML5RMO4RRBLJT6U5","bundle":"https://pith.science/pith/IZQGIGIVLYML5RMO4RRBLJT6U5/bundle.json","state":"https://pith.science/pith/IZQGIGIVLYML5RMO4RRBLJT6U5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IZQGIGIVLYML5RMO4RRBLJT6U5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:IZQGIGIVLYML5RMO4RRBLJT6U5","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":"6fc52ff3de55ebebb848fa1e663a8942a820a2e1719c9885a070501e8495a3cc","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-21T18:41:26Z","title_canon_sha256":"57f77f16e7606d73359dd7bc0c8ac6eac1b8df0d6320e99f2d306c700e3aaf5b"},"schema_version":"1.0","source":{"id":"2311.14743","kind":"arxiv","version":7}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.14743","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"arxiv_version","alias_value":"2311.14743v7","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.14743","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"pith_short_12","alias_value":"IZQGIGIVLYML","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"pith_short_16","alias_value":"IZQGIGIVLYML5RMO","created_at":"2026-07-05T07:36:56Z"},{"alias_kind":"pith_short_8","alias_value":"IZQGIGIV","created_at":"2026-07-05T07:36:56Z"}],"graph_snapshots":[{"event_id":"sha256:7dfec937026c49e10389fe71e82d58a2e228ba84ca27f7874588e4f5a33b5a7a","target":"graph","created_at":"2026-07-05T07:36:56Z","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/2311.14743/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Foundation models, specifically Large Language Models (LLMs), have lately gained wide-spread attention and adoption. Reinforcement Learning with Human Feedback (RLHF) involves training a reward model to capture desired behaviors, which is then used to align LLM's. These reward models are additionally used at inference-time to estimate LLM responses' adherence to those desired behaviors. However, there is little work measuring how robust these reward models are to distribution shifts. In this work, we evaluate how reward model performance - measured via accuracy and calibration (i.e. alignment ","authors_text":"Anthony Chen, Benjamin Pikus, Sean Hendryx, Will LeVine","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-21T18:41:26Z","title":"A Baseline Analysis of Reward Models' Ability To Accurately Analyze Foundation Models Under Distribution Shift"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.14743","kind":"arxiv","version":7},"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:abc9c23df57687555cbae3b9a64689ab23f6ac64be7115f280cb0aaa92a76f5c","target":"record","created_at":"2026-07-05T07:36:56Z","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":"6fc52ff3de55ebebb848fa1e663a8942a820a2e1719c9885a070501e8495a3cc","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-21T18:41:26Z","title_canon_sha256":"57f77f16e7606d73359dd7bc0c8ac6eac1b8df0d6320e99f2d306c700e3aaf5b"},"schema_version":"1.0","source":{"id":"2311.14743","kind":"arxiv","version":7}},"canonical_sha256":"46606419155e18bec58ee46215a67ea75503c1c1810d26751a1e176813b6908a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"46606419155e18bec58ee46215a67ea75503c1c1810d26751a1e176813b6908a","first_computed_at":"2026-07-05T07:36:56.449386Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:36:56.449386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UYZ0XmVaxaPwbBd+wsyJ0iGP0OTXPLr9o4k9u6ZJwUtTQkxRKKPE53sLcXDoI2rU8gb//E8357hFJMduaBPUDA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:36:56.449850Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.14743","source_kind":"arxiv","source_version":7}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:abc9c23df57687555cbae3b9a64689ab23f6ac64be7115f280cb0aaa92a76f5c","sha256:7dfec937026c49e10389fe71e82d58a2e228ba84ca27f7874588e4f5a33b5a7a"],"state_sha256":"c089b56ca7a90f4a52fe04c9094f0d749910068fc03b9853013b2057bf828c6f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yZK99CwIdv/yT5aFG30NMN8shq+qhDR5BYvHtu1ID6UDJZta7pLaKmRIHkhoVIOYoIkGVovF7QBSwEsaTUALAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T09:50:04.126547Z","bundle_sha256":"0d4cd3367fe54beb2124efda3962ab914172ae8beb2de1c9d7ca9eaab2d0e2d3"}}