{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:EESR5EGJT2MVGAXDK6NRJSVSNP","short_pith_number":"pith:EESR5EGJ","canonical_record":{"source":{"id":"2607.01567","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-02T00:53:33Z","cross_cats_sorted":[],"title_canon_sha256":"a85355b3641376327c7789ba918f352c07732f9dd62208d1d1255221760c6b22","abstract_canon_sha256":"b8dce63bba6417fbc4217141b381ea10c40c9b957b34888495cfba0628b7cc01"},"schema_version":"1.0"},"canonical_sha256":"21251e90c99e995302e3579b14cab26be24e94dc606ccf5c32dba664a32bac68","source":{"kind":"arxiv","id":"2607.01567","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.01567","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"arxiv_version","alias_value":"2607.01567v1","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01567","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"pith_short_12","alias_value":"EESR5EGJT2MV","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"pith_short_16","alias_value":"EESR5EGJT2MVGAXD","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"pith_short_8","alias_value":"EESR5EGJ","created_at":"2026-07-03T01:17:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:EESR5EGJT2MVGAXDK6NRJSVSNP","target":"record","payload":{"canonical_record":{"source":{"id":"2607.01567","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-02T00:53:33Z","cross_cats_sorted":[],"title_canon_sha256":"a85355b3641376327c7789ba918f352c07732f9dd62208d1d1255221760c6b22","abstract_canon_sha256":"b8dce63bba6417fbc4217141b381ea10c40c9b957b34888495cfba0628b7cc01"},"schema_version":"1.0"},"canonical_sha256":"21251e90c99e995302e3579b14cab26be24e94dc606ccf5c32dba664a32bac68","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:02.072721Z","signature_b64":"VqIPIe8lRftn603X313VsP5lG92f8cSMFKQrIgv1V5N4/EReNulGvo9tG2aLV33GVsO+1eE2WkmfhPohVFBDAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21251e90c99e995302e3579b14cab26be24e94dc606ccf5c32dba664a32bac68","last_reissued_at":"2026-07-03T01:17:02.072209Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:02.072209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.01567","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-03T01:17:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+uYz2nhz1gMtSSA5wBi0yuNwJa9Crbl0tZZ71vo1sv59uXscGW6tdYZS/LxtMNNV30bczQHYhqqZHaizwlp9Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T18:16:09.682233Z"},"content_sha256":"308a728890477e6c791a982c560bdc2e4ff4f332cbb54177c48c3530482c66f1","schema_version":"1.0","event_id":"sha256:308a728890477e6c791a982c560bdc2e4ff4f332cbb54177c48c3530482c66f1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:EESR5EGJT2MVGAXDK6NRJSVSNP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scaling Trends for Lie Detector Oversight in Preference Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Adam Gleave, Ann-Kathrin Dombrowski, Chris Cundy, Oskar J. Hollinsworth, Sam Adam-Day","submitted_at":"2026-07-02T00:53:33Z","abstract_excerpt":"Deceptive behavior in LLMs is costly to monitor and prevent, motivating approaches such as Scalable Oversight via Lie Detectors (SOLiD) (Cundy & Gleave, 2025), which uses lie detectors to identify responses for review by high-cost labelers. In this paper, we scale SOLiD to larger models and evaluate it in more diverse and realistic preference-learning settings. We find favorable scaling: undetected deception drops from 34% for 1B-parameter models to 14% for 405B-parameter models at a detector true positive rate of 99%, and expensive human labelers can be removed entirely from the fine-tuning p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01567","kind":"arxiv","version":1},"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/2607.01567/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":"2026-07-03T01:17:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Pu+IIIQx7xyAgPV1suSXmSdcCRJAVffN7A+HGSouts4wCGqEV+w7YQw7QY02MOY3kY4CANy/5RMO5NG0nF4rCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T18:16:09.682621Z"},"content_sha256":"a2471faef2d9f04e7ec0f73760d3e19125e02f370b71c45095872a54d4c5b681","schema_version":"1.0","event_id":"sha256:a2471faef2d9f04e7ec0f73760d3e19125e02f370b71c45095872a54d4c5b681"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EESR5EGJT2MVGAXDK6NRJSVSNP/bundle.json","state_url":"https://pith.science/pith/EESR5EGJT2MVGAXDK6NRJSVSNP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EESR5EGJT2MVGAXDK6NRJSVSNP/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":"2026-07-04T18:16:09Z","links":{"resolver":"https://pith.science/pith/EESR5EGJT2MVGAXDK6NRJSVSNP","bundle":"https://pith.science/pith/EESR5EGJT2MVGAXDK6NRJSVSNP/bundle.json","state":"https://pith.science/pith/EESR5EGJT2MVGAXDK6NRJSVSNP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EESR5EGJT2MVGAXDK6NRJSVSNP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EESR5EGJT2MVGAXDK6NRJSVSNP","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":"b8dce63bba6417fbc4217141b381ea10c40c9b957b34888495cfba0628b7cc01","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-02T00:53:33Z","title_canon_sha256":"a85355b3641376327c7789ba918f352c07732f9dd62208d1d1255221760c6b22"},"schema_version":"1.0","source":{"id":"2607.01567","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.01567","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"arxiv_version","alias_value":"2607.01567v1","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01567","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"pith_short_12","alias_value":"EESR5EGJT2MV","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"pith_short_16","alias_value":"EESR5EGJT2MVGAXD","created_at":"2026-07-03T01:17:02Z"},{"alias_kind":"pith_short_8","alias_value":"EESR5EGJ","created_at":"2026-07-03T01:17:02Z"}],"graph_snapshots":[{"event_id":"sha256:a2471faef2d9f04e7ec0f73760d3e19125e02f370b71c45095872a54d4c5b681","target":"graph","created_at":"2026-07-03T01:17:02Z","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/2607.01567/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deceptive behavior in LLMs is costly to monitor and prevent, motivating approaches such as Scalable Oversight via Lie Detectors (SOLiD) (Cundy & Gleave, 2025), which uses lie detectors to identify responses for review by high-cost labelers. In this paper, we scale SOLiD to larger models and evaluate it in more diverse and realistic preference-learning settings. We find favorable scaling: undetected deception drops from 34% for 1B-parameter models to 14% for 405B-parameter models at a detector true positive rate of 99%, and expensive human labelers can be removed entirely from the fine-tuning p","authors_text":"Adam Gleave, Ann-Kathrin Dombrowski, Chris Cundy, Oskar J. Hollinsworth, Sam Adam-Day","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-02T00:53:33Z","title":"Scaling Trends for Lie Detector Oversight in Preference Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01567","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":"sha256:308a728890477e6c791a982c560bdc2e4ff4f332cbb54177c48c3530482c66f1","target":"record","created_at":"2026-07-03T01:17:02Z","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":"b8dce63bba6417fbc4217141b381ea10c40c9b957b34888495cfba0628b7cc01","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-02T00:53:33Z","title_canon_sha256":"a85355b3641376327c7789ba918f352c07732f9dd62208d1d1255221760c6b22"},"schema_version":"1.0","source":{"id":"2607.01567","kind":"arxiv","version":1}},"canonical_sha256":"21251e90c99e995302e3579b14cab26be24e94dc606ccf5c32dba664a32bac68","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"21251e90c99e995302e3579b14cab26be24e94dc606ccf5c32dba664a32bac68","first_computed_at":"2026-07-03T01:17:02.072209Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-03T01:17:02.072209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VqIPIe8lRftn603X313VsP5lG92f8cSMFKQrIgv1V5N4/EReNulGvo9tG2aLV33GVsO+1eE2WkmfhPohVFBDAQ==","signature_status":"signed_v1","signed_at":"2026-07-03T01:17:02.072721Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.01567","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:308a728890477e6c791a982c560bdc2e4ff4f332cbb54177c48c3530482c66f1","sha256:a2471faef2d9f04e7ec0f73760d3e19125e02f370b71c45095872a54d4c5b681"],"state_sha256":"634ad2f470f98638a74caad9915892249bfc3976d8a20476e36e5be23e42238c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TdqAZpvRc9Y9fCKBCLmhj/gmkYF/qKjS0lKTSawxNPfTfmxekiNHcN0iPrPnC9yGlNgg2C49d8uNCvCDaYraDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T18:16:09.684714Z","bundle_sha256":"268c6059bcb14571d2dc26cabc0ddcfdf994d180ebe0fd51a6b4363998e567ee"}}