{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YJP4ZK7PA7D6BI5MROCZDXGWKU","short_pith_number":"pith:YJP4ZK7P","canonical_record":{"source":{"id":"2511.21654","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-11-26T18:27:17Z","cross_cats_sorted":[],"title_canon_sha256":"a463c37c9d1a2753bf460aac698e88213587f33b55cf78a6391489e67ca45401","abstract_canon_sha256":"2b8c191206447764c56d6e629bc450640b46cb4c64e80e390193239369fb8408"},"schema_version":"1.0"},"canonical_sha256":"c25fccabef07c7e0a3ac8b8591dcd6553989a032cfdc8a89b24fd9b7f427e82c","source":{"kind":"arxiv","id":"2511.21654","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.21654","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2511.21654v2","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.21654","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"YJP4ZK7PA7D6","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"YJP4ZK7PA7D6BI5M","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"YJP4ZK7P","created_at":"2026-05-20T00:04:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YJP4ZK7PA7D6BI5MROCZDXGWKU","target":"record","payload":{"canonical_record":{"source":{"id":"2511.21654","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-11-26T18:27:17Z","cross_cats_sorted":[],"title_canon_sha256":"a463c37c9d1a2753bf460aac698e88213587f33b55cf78a6391489e67ca45401","abstract_canon_sha256":"2b8c191206447764c56d6e629bc450640b46cb4c64e80e390193239369fb8408"},"schema_version":"1.0"},"canonical_sha256":"c25fccabef07c7e0a3ac8b8591dcd6553989a032cfdc8a89b24fd9b7f427e82c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:17.929336Z","signature_b64":"MRbbvdUKTXvSzQEipreYcEjxEgDIxJf1AldVeYEj/fUdo8apvCdC7supuQhAFE/SMEJL8We9V5/d3yAyIstyBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c25fccabef07c7e0a3ac8b8591dcd6553989a032cfdc8a89b24fd9b7f427e82c","last_reissued_at":"2026-05-20T00:04:17.928534Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:17.928534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2511.21654","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6+2WlTbIr5SnWjfRW1o9yy+7xOudm0AnDxVRrdzmmTGBrCIbsRQIP8CFbeqJY6eSt1AYJVfNkGeT/NOOhYAUCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T18:40:12.389944Z"},"content_sha256":"91cb6d8ad9214bb19482c9b46fa4460c89d33c6d80b248531d508951f5cd0914","schema_version":"1.0","event_id":"sha256:91cb6d8ad9214bb19482c9b46fa4460c89d33c6d80b248531d508951f5cd0914"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YJP4ZK7PA7D6BI5MROCZDXGWKU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EvilGenie: A Reward Hacking Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jayson Lynch, Jonathan Gabor, Jonathan Rosenfeld","submitted_at":"2025-11-26T18:27:17Z","abstract_excerpt":"We introduce EvilGenie, a benchmark for reward hacking in programming settings. We source problems from LiveCodeBench and create an environment in which agents can easily reward hack, such as by hardcoding test cases or editing the testing files. We measure reward hacking in three ways: held out unit tests, LLM judges, and test file edit detection. We verify these methods against human review and each other. We find the LLM judge to be highly effective at detecting reward hacking in unambiguous cases, and observe only minimal improvement from the use of held out test cases. In addition to test"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.21654","kind":"arxiv","version":2},"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/2511.21654/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":"2026-05-20T00:04:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u+zmqdSxXlm0GP9VNhJif4jujX/UjmCu6nQNhj66kEGjw6W6VcNDtUkNNH/rfkWJYk4Hdx5B8xfTQZ4wn4RLBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T18:40:12.390648Z"},"content_sha256":"2d5af73384e3247f668fa8f2331254ffca5502e209e943af5bd97b4db8f6b083","schema_version":"1.0","event_id":"sha256:2d5af73384e3247f668fa8f2331254ffca5502e209e943af5bd97b4db8f6b083"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU/bundle.json","state_url":"https://pith.science/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU/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":"2026-05-27T18:40:12Z","links":{"resolver":"https://pith.science/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU","bundle":"https://pith.science/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU/bundle.json","state":"https://pith.science/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YJP4ZK7PA7D6BI5MROCZDXGWKU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YJP4ZK7PA7D6BI5MROCZDXGWKU","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":"2b8c191206447764c56d6e629bc450640b46cb4c64e80e390193239369fb8408","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-11-26T18:27:17Z","title_canon_sha256":"a463c37c9d1a2753bf460aac698e88213587f33b55cf78a6391489e67ca45401"},"schema_version":"1.0","source":{"id":"2511.21654","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.21654","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2511.21654v2","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.21654","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"YJP4ZK7PA7D6","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"YJP4ZK7PA7D6BI5M","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"YJP4ZK7P","created_at":"2026-05-20T00:04:17Z"}],"graph_snapshots":[{"event_id":"sha256:2d5af73384e3247f668fa8f2331254ffca5502e209e943af5bd97b4db8f6b083","target":"graph","created_at":"2026-05-20T00:04:17Z","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/2511.21654/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce EvilGenie, a benchmark for reward hacking in programming settings. We source problems from LiveCodeBench and create an environment in which agents can easily reward hack, such as by hardcoding test cases or editing the testing files. We measure reward hacking in three ways: held out unit tests, LLM judges, and test file edit detection. We verify these methods against human review and each other. We find the LLM judge to be highly effective at detecting reward hacking in unambiguous cases, and observe only minimal improvement from the use of held out test cases. In addition to test","authors_text":"Jayson Lynch, Jonathan Gabor, Jonathan Rosenfeld","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-11-26T18:27:17Z","title":"EvilGenie: A Reward Hacking Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.21654","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":"sha256:91cb6d8ad9214bb19482c9b46fa4460c89d33c6d80b248531d508951f5cd0914","target":"record","created_at":"2026-05-20T00:04:17Z","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":"2b8c191206447764c56d6e629bc450640b46cb4c64e80e390193239369fb8408","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-11-26T18:27:17Z","title_canon_sha256":"a463c37c9d1a2753bf460aac698e88213587f33b55cf78a6391489e67ca45401"},"schema_version":"1.0","source":{"id":"2511.21654","kind":"arxiv","version":2}},"canonical_sha256":"c25fccabef07c7e0a3ac8b8591dcd6553989a032cfdc8a89b24fd9b7f427e82c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c25fccabef07c7e0a3ac8b8591dcd6553989a032cfdc8a89b24fd9b7f427e82c","first_computed_at":"2026-05-20T00:04:17.928534Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:17.928534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MRbbvdUKTXvSzQEipreYcEjxEgDIxJf1AldVeYEj/fUdo8apvCdC7supuQhAFE/SMEJL8We9V5/d3yAyIstyBQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:17.929336Z","signed_message":"canonical_sha256_bytes"},"source_id":"2511.21654","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:91cb6d8ad9214bb19482c9b46fa4460c89d33c6d80b248531d508951f5cd0914","sha256:2d5af73384e3247f668fa8f2331254ffca5502e209e943af5bd97b4db8f6b083"],"state_sha256":"255e4182ce5750fb1bb625f6c87e75298c1421dfa97f8f418bea9e6c400ad75a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"495ifuuNbaLmPu+wP8DS6UVqLyviKOqIZXwlPr95Raz8cSBmHJeSR4AI2MCruUzwzLdOw+ni4tPARDlab/o0BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T18:40:12.393927Z","bundle_sha256":"170c0185c5450b3505f3e19a13deaa66c162e4b2c6421adf780f029d4a649357"}}