{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:N6WXARGNMN7LB5AT4MHSIHDOBG","short_pith_number":"pith:N6WXARGN","canonical_record":{"source":{"id":"2607.20852","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T02:23:09Z","cross_cats_sorted":[],"title_canon_sha256":"4fa0eb9e60930fa0a01fe666e9ba941cbef5a7529653b71140b835d6af0b9a4d","abstract_canon_sha256":"f4a2455c81dce7aa5c175e326b3beffbe74b9bb41e07da951fe906182a773e0d"},"schema_version":"1.0"},"canonical_sha256":"6fad7044cd637eb0f413e30f241c6e0981406611127011d69818adf70a6c2cb0","source":{"kind":"arxiv","id":"2607.20852","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.20852","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"arxiv_version","alias_value":"2607.20852v1","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20852","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"pith_short_12","alias_value":"N6WXARGNMN7L","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"pith_short_16","alias_value":"N6WXARGNMN7LB5AT","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"pith_short_8","alias_value":"N6WXARGN","created_at":"2026-07-24T00:23:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:N6WXARGNMN7LB5AT4MHSIHDOBG","target":"record","payload":{"canonical_record":{"source":{"id":"2607.20852","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T02:23:09Z","cross_cats_sorted":[],"title_canon_sha256":"4fa0eb9e60930fa0a01fe666e9ba941cbef5a7529653b71140b835d6af0b9a4d","abstract_canon_sha256":"f4a2455c81dce7aa5c175e326b3beffbe74b9bb41e07da951fe906182a773e0d"},"schema_version":"1.0"},"canonical_sha256":"6fad7044cd637eb0f413e30f241c6e0981406611127011d69818adf70a6c2cb0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T00:23:38.190366Z","signature_b64":"bJ3fOFIvc4JkCxpELKZ6SXWq0rSQ7fotVo/WGmd40/jI9veilxylRQ3bnESZC2wNKKVfq2/8JlZL7rmWwpFVDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6fad7044cd637eb0f413e30f241c6e0981406611127011d69818adf70a6c2cb0","last_reissued_at":"2026-07-24T00:23:38.189470Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T00:23:38.189470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.20852","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":"2026-07-24T00:23:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZDaKMAGcUInMWeGH6BeOHvtP7OYIMFGYhGSH2ep0pg07xe8o0/aE4soFnmrI5H4eV6LyKvkeCgD6F0jbpLwOCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:37:30.452940Z"},"content_sha256":"667d316ea07a883ef5ff1bd6a16d78f627c3d7c08cba08b5150dbda7dcea8fb7","schema_version":"1.0","event_id":"sha256:667d316ea07a883ef5ff1bd6a16d78f627c3d7c08cba08b5150dbda7dcea8fb7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:N6WXARGNMN7LB5AT4MHSIHDOBG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Code Monitor Red Teaming for Public-Test-Passing Code","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fuji Ren, Jiawen Deng, Junchi Liao","submitted_at":"2026-07-23T02:23:09Z","abstract_excerpt":"Visible tests are a common gate for LLM-generated code, but passing them does not certify specification correctness. We study a deployment-like monitoring problem: after code has passed public tests, can a weaker LLM verifier identify the residual hidden bugs? We introduce Code Monitor Red Teaming, a monitor-red-teaming protocol that fixes a public-check information boundary while varying generator pressure, verifier scaffolding, and weak-to-strong capability. We instantiate it as CodeMonitorBench, spanning function-level, data-science, and workflow code. Across 71,000 generated candidates, 43"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20852","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/2607.20852/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":"2026-07-24T00:23:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RaG2UIShUD43EgaLM3RMpds3dZALgMPbG6wSxouYx3+IV5fs4QnYjQIVw2q022ip6nUv7ywK4GFy3qMuVakrDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:37:30.453447Z"},"content_sha256":"b5cf700e226f0b5bea116cc58540e91039721d5b2a970c04d9f853b7d1e66907","schema_version":"1.0","event_id":"sha256:b5cf700e226f0b5bea116cc58540e91039721d5b2a970c04d9f853b7d1e66907"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N6WXARGNMN7LB5AT4MHSIHDOBG/bundle.json","state_url":"https://pith.science/pith/N6WXARGNMN7LB5AT4MHSIHDOBG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N6WXARGNMN7LB5AT4MHSIHDOBG/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":"2026-08-04T12:37:30Z","links":{"resolver":"https://pith.science/pith/N6WXARGNMN7LB5AT4MHSIHDOBG","bundle":"https://pith.science/pith/N6WXARGNMN7LB5AT4MHSIHDOBG/bundle.json","state":"https://pith.science/pith/N6WXARGNMN7LB5AT4MHSIHDOBG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N6WXARGNMN7LB5AT4MHSIHDOBG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:N6WXARGNMN7LB5AT4MHSIHDOBG","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":"f4a2455c81dce7aa5c175e326b3beffbe74b9bb41e07da951fe906182a773e0d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T02:23:09Z","title_canon_sha256":"4fa0eb9e60930fa0a01fe666e9ba941cbef5a7529653b71140b835d6af0b9a4d"},"schema_version":"1.0","source":{"id":"2607.20852","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.20852","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"arxiv_version","alias_value":"2607.20852v1","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20852","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"pith_short_12","alias_value":"N6WXARGNMN7L","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"pith_short_16","alias_value":"N6WXARGNMN7LB5AT","created_at":"2026-07-24T00:23:38Z"},{"alias_kind":"pith_short_8","alias_value":"N6WXARGN","created_at":"2026-07-24T00:23:38Z"}],"graph_snapshots":[{"event_id":"sha256:b5cf700e226f0b5bea116cc58540e91039721d5b2a970c04d9f853b7d1e66907","target":"graph","created_at":"2026-07-24T00:23:38Z","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/2607.20852/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Visible tests are a common gate for LLM-generated code, but passing them does not certify specification correctness. We study a deployment-like monitoring problem: after code has passed public tests, can a weaker LLM verifier identify the residual hidden bugs? We introduce Code Monitor Red Teaming, a monitor-red-teaming protocol that fixes a public-check information boundary while varying generator pressure, verifier scaffolding, and weak-to-strong capability. We instantiate it as CodeMonitorBench, spanning function-level, data-science, and workflow code. Across 71,000 generated candidates, 43","authors_text":"Fuji Ren, Jiawen Deng, Junchi Liao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T02:23:09Z","title":"Code Monitor Red Teaming for Public-Test-Passing Code"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20852","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":"sha256:667d316ea07a883ef5ff1bd6a16d78f627c3d7c08cba08b5150dbda7dcea8fb7","target":"record","created_at":"2026-07-24T00:23:38Z","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":"f4a2455c81dce7aa5c175e326b3beffbe74b9bb41e07da951fe906182a773e0d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T02:23:09Z","title_canon_sha256":"4fa0eb9e60930fa0a01fe666e9ba941cbef5a7529653b71140b835d6af0b9a4d"},"schema_version":"1.0","source":{"id":"2607.20852","kind":"arxiv","version":1}},"canonical_sha256":"6fad7044cd637eb0f413e30f241c6e0981406611127011d69818adf70a6c2cb0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6fad7044cd637eb0f413e30f241c6e0981406611127011d69818adf70a6c2cb0","first_computed_at":"2026-07-24T00:23:38.189470Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-24T00:23:38.189470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bJ3fOFIvc4JkCxpELKZ6SXWq0rSQ7fotVo/WGmd40/jI9veilxylRQ3bnESZC2wNKKVfq2/8JlZL7rmWwpFVDg==","signature_status":"signed_v1","signed_at":"2026-07-24T00:23:38.190366Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.20852","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:667d316ea07a883ef5ff1bd6a16d78f627c3d7c08cba08b5150dbda7dcea8fb7","sha256:b5cf700e226f0b5bea116cc58540e91039721d5b2a970c04d9f853b7d1e66907"],"state_sha256":"8ea015430b8cd314382681a0878b0b86420f91ec623fe34da3b5cd3d077b3ae1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HUa7feSk7ExZiVIeWaMKzxHiU9R5Ro6+IPS11W5CcmMPVFjFdSTF8n9VAQkZX2PaCp5mNWxVpMRPWqQuBSbiAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T12:37:30.457775Z","bundle_sha256":"21d865a9e4bee4ad7cde86226b2535fdefd85a700ec7321589ad6f03fa9110a9"}}