{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:2TJWZ33WLT2KKKTXFLNYBQLUDP","short_pith_number":"pith:2TJWZ33W","canonical_record":{"source":{"id":"2507.03010","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T10:48:37Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"52b67495c80dc431b214140be475038d03bb1b9e4fe74170d9fbaa69572e9c6b","abstract_canon_sha256":"f90215c0cd56f8e955313207252e9f365028951ea31f9760cee097256867c2ac"},"schema_version":"1.0"},"canonical_sha256":"d4d36cef765cf4a52a772adb80c1741bd9777934f5842bfa153781b21a91cfba","source":{"kind":"arxiv","id":"2507.03010","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.03010","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"arxiv_version","alias_value":"2507.03010v1","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03010","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"pith_short_12","alias_value":"2TJWZ33WLT2K","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"pith_short_16","alias_value":"2TJWZ33WLT2KKKTX","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"pith_short_8","alias_value":"2TJWZ33W","created_at":"2026-07-05T11:31:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:2TJWZ33WLT2KKKTXFLNYBQLUDP","target":"record","payload":{"canonical_record":{"source":{"id":"2507.03010","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T10:48:37Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"52b67495c80dc431b214140be475038d03bb1b9e4fe74170d9fbaa69572e9c6b","abstract_canon_sha256":"f90215c0cd56f8e955313207252e9f365028951ea31f9760cee097256867c2ac"},"schema_version":"1.0"},"canonical_sha256":"d4d36cef765cf4a52a772adb80c1741bd9777934f5842bfa153781b21a91cfba","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:50.635134Z","signature_b64":"JuXyQg43Foky5luauHEeH6+7G2kfnJTMvD9zudajkCtKieSLAPPzgd/xzl8UoVolmrWYhBpvm+Fw/cTv6saEAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4d36cef765cf4a52a772adb80c1741bd9777934f5842bfa153781b21a91cfba","last_reissued_at":"2026-07-05T11:31:50.634673Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:50.634673Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.03010","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":"2026-07-05T11:31:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I2txAXEuCNmsI3im+9PmT8ID761f92WHBHXF94c+7HxQ3ZwAyodWwrbfuCWZX3BR98pthU7SRNpoZpcGsEw+CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:34:16.045135Z"},"content_sha256":"730fca2fc9c7a4acfd505ee90d166ad6b92d51970310dff1fa08e1745000eea2","schema_version":"1.0","event_id":"sha256:730fca2fc9c7a4acfd505ee90d166ad6b92d51970310dff1fa08e1745000eea2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:2TJWZ33WLT2KKKTXFLNYBQLUDP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Subversion via Focal Points: Investigating Collusion in LLM Monitoring","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.CL","authors_text":"Olli J\\\"arviniemi","submitted_at":"2025-07-02T10:48:37Z","abstract_excerpt":"We evaluate language models' ability to subvert monitoring protocols via collusion. More specifically, we have two instances of a model design prompts for a policy (P) and a monitor (M) in a programming task setting. The models collaboratively aim for M to classify all backdoored programs in an auditing dataset as harmful, but nevertheless classify a backdoored program produced by P as harmless. The models are isolated from each other, requiring them to independently arrive at compatible subversion strategies. We find that while Claude 3.7 Sonnet has low success rate due to poor convergence, i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03010","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/2507.03010/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":"2026-07-05T11:31:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6peg23LdS3QPBP6s4wgIJ6p6YFCNKsUNarETfrl+OmPew3wzYfuncpV3uWrYhT2YXDxb8lT2UbS0zzMfTTi8Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:34:16.045710Z"},"content_sha256":"39eecc93220ecf2cb75c710b2c1155760a64cae2012ea762e6d79b66340229f6","schema_version":"1.0","event_id":"sha256:39eecc93220ecf2cb75c710b2c1155760a64cae2012ea762e6d79b66340229f6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP/bundle.json","state_url":"https://pith.science/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP/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":"2026-08-07T18:34:16Z","links":{"resolver":"https://pith.science/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP","bundle":"https://pith.science/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP/bundle.json","state":"https://pith.science/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2TJWZ33WLT2KKKTXFLNYBQLUDP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:2TJWZ33WLT2KKKTXFLNYBQLUDP","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":"f90215c0cd56f8e955313207252e9f365028951ea31f9760cee097256867c2ac","cross_cats_sorted":["cs.CR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T10:48:37Z","title_canon_sha256":"52b67495c80dc431b214140be475038d03bb1b9e4fe74170d9fbaa69572e9c6b"},"schema_version":"1.0","source":{"id":"2507.03010","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.03010","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"arxiv_version","alias_value":"2507.03010v1","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03010","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"pith_short_12","alias_value":"2TJWZ33WLT2K","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"pith_short_16","alias_value":"2TJWZ33WLT2KKKTX","created_at":"2026-07-05T11:31:50Z"},{"alias_kind":"pith_short_8","alias_value":"2TJWZ33W","created_at":"2026-07-05T11:31:50Z"}],"graph_snapshots":[{"event_id":"sha256:39eecc93220ecf2cb75c710b2c1155760a64cae2012ea762e6d79b66340229f6","target":"graph","created_at":"2026-07-05T11:31:50Z","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/2507.03010/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We evaluate language models' ability to subvert monitoring protocols via collusion. More specifically, we have two instances of a model design prompts for a policy (P) and a monitor (M) in a programming task setting. The models collaboratively aim for M to classify all backdoored programs in an auditing dataset as harmful, but nevertheless classify a backdoored program produced by P as harmless. The models are isolated from each other, requiring them to independently arrive at compatible subversion strategies. We find that while Claude 3.7 Sonnet has low success rate due to poor convergence, i","authors_text":"Olli J\\\"arviniemi","cross_cats":["cs.CR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T10:48:37Z","title":"Subversion via Focal Points: Investigating Collusion in LLM Monitoring"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03010","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":"sha256:730fca2fc9c7a4acfd505ee90d166ad6b92d51970310dff1fa08e1745000eea2","target":"record","created_at":"2026-07-05T11:31:50Z","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":"f90215c0cd56f8e955313207252e9f365028951ea31f9760cee097256867c2ac","cross_cats_sorted":["cs.CR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T10:48:37Z","title_canon_sha256":"52b67495c80dc431b214140be475038d03bb1b9e4fe74170d9fbaa69572e9c6b"},"schema_version":"1.0","source":{"id":"2507.03010","kind":"arxiv","version":1}},"canonical_sha256":"d4d36cef765cf4a52a772adb80c1741bd9777934f5842bfa153781b21a91cfba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d4d36cef765cf4a52a772adb80c1741bd9777934f5842bfa153781b21a91cfba","first_computed_at":"2026-07-05T11:31:50.634673Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:31:50.634673Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JuXyQg43Foky5luauHEeH6+7G2kfnJTMvD9zudajkCtKieSLAPPzgd/xzl8UoVolmrWYhBpvm+Fw/cTv6saEAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:31:50.635134Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.03010","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:730fca2fc9c7a4acfd505ee90d166ad6b92d51970310dff1fa08e1745000eea2","sha256:39eecc93220ecf2cb75c710b2c1155760a64cae2012ea762e6d79b66340229f6"],"state_sha256":"f8052b38b44461412fae405f50b20cc1769bcc950be6e7210097ead9c73ea095"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2KsBRwSPQZT90hdfqqHDN8aooR0XLrSxLON8/u357OPq1Nvr44lPyvJZVYP8cU/C10yEoqaCF4lkNF/7X6d0Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T18:34:16.049709Z","bundle_sha256":"2f511276218d3c6a4dfc476d847a5fb6baf757f47d0ff64501fb5515f3af47b8"}}