{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:NRI6OEDOGBDPYWFSMQRFGGKTRX","short_pith_number":"pith:NRI6OEDO","canonical_record":{"source":{"id":"2402.06627","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T18:59:29Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"7b37f06c0552076f15c256e032c946bf294cc507e6460df0c13c0f351b999d6d","abstract_canon_sha256":"7d3f5aacb68a93a40c5d5d9249fc970fade483a6398c897fb34cb60121aeb983"},"schema_version":"1.0"},"canonical_sha256":"6c51e7106e3046fc58b264225319538ddd8cc9e9f29b2deea091b3493503a5b4","source":{"kind":"arxiv","id":"2402.06627","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.06627","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"arxiv_version","alias_value":"2402.06627v3","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06627","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_12","alias_value":"NRI6OEDOGBDP","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_16","alias_value":"NRI6OEDOGBDPYWFS","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_8","alias_value":"NRI6OEDO","created_at":"2026-07-05T08:28:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:NRI6OEDOGBDPYWFSMQRFGGKTRX","target":"record","payload":{"canonical_record":{"source":{"id":"2402.06627","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T18:59:29Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"7b37f06c0552076f15c256e032c946bf294cc507e6460df0c13c0f351b999d6d","abstract_canon_sha256":"7d3f5aacb68a93a40c5d5d9249fc970fade483a6398c897fb34cb60121aeb983"},"schema_version":"1.0"},"canonical_sha256":"6c51e7106e3046fc58b264225319538ddd8cc9e9f29b2deea091b3493503a5b4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:37.979172Z","signature_b64":"BOY6m0CBfh/euhejS+nhhaFoBd1bXoUWktq9/d9M4WROLW1rWL7RNE9oYrpCr3OzY0PT+LlNyPe2/xMPCVd4CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c51e7106e3046fc58b264225319538ddd8cc9e9f29b2deea091b3493503a5b4","last_reissued_at":"2026-07-05T08:28:37.978707Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:37.978707Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.06627","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:28:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7NlXADU0ANYZLbzG8ZwnLOEysarTjopZxaswWKhZjD66soJndlDlkmDdyL34ElBwDrrcpQJ6BqMcE4ZejUS6Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T07:19:37.116046Z"},"content_sha256":"19708705c51be7c8a68ffaacc9adc130da88f1515ee641cfc4474c6e55a25770","schema_version":"1.0","event_id":"sha256:19708705c51be7c8a68ffaacc9adc130da88f1515ee641cfc4474c6e55a25770"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:NRI6OEDOGBDPYWFSMQRFGGKTRX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Feedback Loops With Language Models Drive In-Context Reward Hacking","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Alexander Pan, Erik Jones, Jacob Steinhardt, Meena Jagadeesan","submitted_at":"2024-02-09T18:59:29Z","abstract_excerpt":"Language models influence the external world: they query APIs that read and write to web pages, generate content that shapes human behavior, and run system commands as autonomous agents. These interactions form feedback loops: LLM outputs affect the world, which in turn affect subsequent LLM outputs. In this work, we show that feedback loops can cause in-context reward hacking (ICRH), where the LLM at test-time optimizes a (potentially implicit) objective but creates negative side effects in the process. For example, consider an LLM agent deployed to increase Twitter engagement; the LLM may re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06627","kind":"arxiv","version":3},"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/2402.06627/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":"2026-07-05T08:28:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DYcZu4ZFGUSw3/fwKrWaxAQ9kZrPCcFYIKn8EyzxKjp9moYlbSME8zgZToJjglWQb3FBZAyJfes5pIYaJUzOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T07:19:37.116564Z"},"content_sha256":"bf5763e573ac71e671d6ab1b8d2abcb972b47893b8466e308089f0ef38220dfc","schema_version":"1.0","event_id":"sha256:bf5763e573ac71e671d6ab1b8d2abcb972b47893b8466e308089f0ef38220dfc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX/bundle.json","state_url":"https://pith.science/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX/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":"2026-08-06T07:19:37Z","links":{"resolver":"https://pith.science/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX","bundle":"https://pith.science/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX/bundle.json","state":"https://pith.science/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NRI6OEDOGBDPYWFSMQRFGGKTRX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NRI6OEDOGBDPYWFSMQRFGGKTRX","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":"7d3f5aacb68a93a40c5d5d9249fc970fade483a6398c897fb34cb60121aeb983","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T18:59:29Z","title_canon_sha256":"7b37f06c0552076f15c256e032c946bf294cc507e6460df0c13c0f351b999d6d"},"schema_version":"1.0","source":{"id":"2402.06627","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.06627","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"arxiv_version","alias_value":"2402.06627v3","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06627","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_12","alias_value":"NRI6OEDOGBDP","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_16","alias_value":"NRI6OEDOGBDPYWFS","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_8","alias_value":"NRI6OEDO","created_at":"2026-07-05T08:28:37Z"}],"graph_snapshots":[{"event_id":"sha256:bf5763e573ac71e671d6ab1b8d2abcb972b47893b8466e308089f0ef38220dfc","target":"graph","created_at":"2026-07-05T08:28:37Z","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/2402.06627/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models influence the external world: they query APIs that read and write to web pages, generate content that shapes human behavior, and run system commands as autonomous agents. These interactions form feedback loops: LLM outputs affect the world, which in turn affect subsequent LLM outputs. In this work, we show that feedback loops can cause in-context reward hacking (ICRH), where the LLM at test-time optimizes a (potentially implicit) objective but creates negative side effects in the process. For example, consider an LLM agent deployed to increase Twitter engagement; the LLM may re","authors_text":"Alexander Pan, Erik Jones, Jacob Steinhardt, Meena Jagadeesan","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T18:59:29Z","title":"Feedback Loops With Language Models Drive In-Context Reward Hacking"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06627","kind":"arxiv","version":3},"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":"sha256:19708705c51be7c8a68ffaacc9adc130da88f1515ee641cfc4474c6e55a25770","target":"record","created_at":"2026-07-05T08:28:37Z","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":"7d3f5aacb68a93a40c5d5d9249fc970fade483a6398c897fb34cb60121aeb983","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T18:59:29Z","title_canon_sha256":"7b37f06c0552076f15c256e032c946bf294cc507e6460df0c13c0f351b999d6d"},"schema_version":"1.0","source":{"id":"2402.06627","kind":"arxiv","version":3}},"canonical_sha256":"6c51e7106e3046fc58b264225319538ddd8cc9e9f29b2deea091b3493503a5b4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c51e7106e3046fc58b264225319538ddd8cc9e9f29b2deea091b3493503a5b4","first_computed_at":"2026-07-05T08:28:37.978707Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:37.978707Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BOY6m0CBfh/euhejS+nhhaFoBd1bXoUWktq9/d9M4WROLW1rWL7RNE9oYrpCr3OzY0PT+LlNyPe2/xMPCVd4CA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:37.979172Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.06627","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:19708705c51be7c8a68ffaacc9adc130da88f1515ee641cfc4474c6e55a25770","sha256:bf5763e573ac71e671d6ab1b8d2abcb972b47893b8466e308089f0ef38220dfc"],"state_sha256":"6903c052c25c6f0c4a9caba1f02e284df4d831ad8ab10a946d9797b49a7dea69"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ae0nNLzwZmIKPdaEFeXRISM2r7sWD1FSQB7N2k14wuNtC5r8h0VKN96Z36wXud9Rrw2Yz8I3qR1OHOc4uovuDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T07:19:37.121668Z","bundle_sha256":"7f544ae8fbc87ec440f08665a0ce6ce1b5cdf78b0035d04594b33d016a995564"}}