{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:TALSN35V477O47NPXPOHWWWGB2","short_pith_number":"pith:TALSN35V","canonical_record":{"source":{"id":"2302.10894","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-08T02:30:07Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"66ed8f1375194d5cc9e4ed4b1d840e0502b6988cf298662952a02f79a3ede6f4","abstract_canon_sha256":"af48894d87dfdaf4690bc0849f340f6f8b7f2564a5280890ae0b2c2983f18529"},"schema_version":"1.0"},"canonical_sha256":"981726efb5e7feee7dafbbdc7b5ac60eaad447c2787942ecac310235961ee7ab","source":{"kind":"arxiv","id":"2302.10894","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.10894","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"arxiv_version","alias_value":"2302.10894v3","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.10894","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"pith_short_12","alias_value":"TALSN35V477O","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"pith_short_16","alias_value":"TALSN35V477O47NP","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"pith_short_8","alias_value":"TALSN35V","created_at":"2026-07-05T06:53:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:TALSN35V477O47NPXPOHWWWGB2","target":"record","payload":{"canonical_record":{"source":{"id":"2302.10894","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-08T02:30:07Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"66ed8f1375194d5cc9e4ed4b1d840e0502b6988cf298662952a02f79a3ede6f4","abstract_canon_sha256":"af48894d87dfdaf4690bc0849f340f6f8b7f2564a5280890ae0b2c2983f18529"},"schema_version":"1.0"},"canonical_sha256":"981726efb5e7feee7dafbbdc7b5ac60eaad447c2787942ecac310235961ee7ab","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:53:05.562141Z","signature_b64":"8QN6NEhrUpylI8+Em67mFhBih+bJMofAhIgP7mMF8dcVkcUXzwTszDiM5woYs2jSTunPE6X1Kv1fO4/fuRSfAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"981726efb5e7feee7dafbbdc7b5ac60eaad447c2787942ecac310235961ee7ab","last_reissued_at":"2026-07-05T06:53:05.561671Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:53:05.561671Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2302.10894","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":"2026-07-05T06:53:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zfhviN4iVffO4jUU7E5QqEfTFgpHvzReJKify08uljd+DRRNcuTsexudH7ZsYqvT7VoaceymfHrdcJpybC63Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T15:23:43.218117Z"},"content_sha256":"b858bb21981aa97e9d39b918227740f8d6edaab5f65e2ace9461131a74155f26","schema_version":"1.0","event_id":"sha256:b858bb21981aa97e9d39b918227740f8d6edaab5f65e2ace9461131a74155f26"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:TALSN35V477O47NPXPOHWWWGB2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Red Teaming Deep Neural Networks with Feature Synthesis Tools","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Dylan Hadfield-Menell, Jiawei Li, Kaivalya Hariharan, Kevin Zhang, Stephen Casper, Tong Bu, Yuxiao Li","submitted_at":"2023-02-08T02:30:07Z","abstract_excerpt":"Interpretable AI tools are often motivated by the goal of understanding model behavior in out-of-distribution (OOD) contexts. Despite the attention this area of study receives, there are comparatively few cases where these tools have identified previously unknown bugs in models. We argue that this is due, in part, to a common feature of many interpretability methods: they analyze model behavior by using a particular dataset. This only allows for the study of the model in the context of features that the user can sample in advance. To address this, a growing body of research involves interpreti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.10894","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/2302.10894/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":"2026-07-05T06:53:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5aZEqukZie8SchyG60MD4AvXONXcVzbRcw2IT0hjArQ2gy5qwXLRTaQe4W54kB4J+HWqkC+xGcMly5+SrvCOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T15:23:43.223531Z"},"content_sha256":"fb82ad0b3d42be714414d18ebbd94d8eba600f461217108eb6bb48b40c028844","schema_version":"1.0","event_id":"sha256:fb82ad0b3d42be714414d18ebbd94d8eba600f461217108eb6bb48b40c028844"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TALSN35V477O47NPXPOHWWWGB2/bundle.json","state_url":"https://pith.science/pith/TALSN35V477O47NPXPOHWWWGB2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TALSN35V477O47NPXPOHWWWGB2/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":"2026-08-13T15:23:43Z","links":{"resolver":"https://pith.science/pith/TALSN35V477O47NPXPOHWWWGB2","bundle":"https://pith.science/pith/TALSN35V477O47NPXPOHWWWGB2/bundle.json","state":"https://pith.science/pith/TALSN35V477O47NPXPOHWWWGB2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TALSN35V477O47NPXPOHWWWGB2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:TALSN35V477O47NPXPOHWWWGB2","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":"af48894d87dfdaf4690bc0849f340f6f8b7f2564a5280890ae0b2c2983f18529","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-08T02:30:07Z","title_canon_sha256":"66ed8f1375194d5cc9e4ed4b1d840e0502b6988cf298662952a02f79a3ede6f4"},"schema_version":"1.0","source":{"id":"2302.10894","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.10894","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"arxiv_version","alias_value":"2302.10894v3","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.10894","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"pith_short_12","alias_value":"TALSN35V477O","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"pith_short_16","alias_value":"TALSN35V477O47NP","created_at":"2026-07-05T06:53:05Z"},{"alias_kind":"pith_short_8","alias_value":"TALSN35V","created_at":"2026-07-05T06:53:05Z"}],"graph_snapshots":[{"event_id":"sha256:fb82ad0b3d42be714414d18ebbd94d8eba600f461217108eb6bb48b40c028844","target":"graph","created_at":"2026-07-05T06:53:05Z","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/2302.10894/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Interpretable AI tools are often motivated by the goal of understanding model behavior in out-of-distribution (OOD) contexts. Despite the attention this area of study receives, there are comparatively few cases where these tools have identified previously unknown bugs in models. We argue that this is due, in part, to a common feature of many interpretability methods: they analyze model behavior by using a particular dataset. This only allows for the study of the model in the context of features that the user can sample in advance. To address this, a growing body of research involves interpreti","authors_text":"Dylan Hadfield-Menell, Jiawei Li, Kaivalya Hariharan, Kevin Zhang, Stephen Casper, Tong Bu, Yuxiao Li","cross_cats":["cs.AI","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-08T02:30:07Z","title":"Red Teaming Deep Neural Networks with Feature Synthesis Tools"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.10894","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:b858bb21981aa97e9d39b918227740f8d6edaab5f65e2ace9461131a74155f26","target":"record","created_at":"2026-07-05T06:53:05Z","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":"af48894d87dfdaf4690bc0849f340f6f8b7f2564a5280890ae0b2c2983f18529","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-08T02:30:07Z","title_canon_sha256":"66ed8f1375194d5cc9e4ed4b1d840e0502b6988cf298662952a02f79a3ede6f4"},"schema_version":"1.0","source":{"id":"2302.10894","kind":"arxiv","version":3}},"canonical_sha256":"981726efb5e7feee7dafbbdc7b5ac60eaad447c2787942ecac310235961ee7ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"981726efb5e7feee7dafbbdc7b5ac60eaad447c2787942ecac310235961ee7ab","first_computed_at":"2026-07-05T06:53:05.561671Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:53:05.561671Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8QN6NEhrUpylI8+Em67mFhBih+bJMofAhIgP7mMF8dcVkcUXzwTszDiM5woYs2jSTunPE6X1Kv1fO4/fuRSfAg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:53:05.562141Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.10894","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b858bb21981aa97e9d39b918227740f8d6edaab5f65e2ace9461131a74155f26","sha256:fb82ad0b3d42be714414d18ebbd94d8eba600f461217108eb6bb48b40c028844"],"state_sha256":"4df577bd7eeef51d3a81e8174529d6ae8f274db5d1c64dabcdb5d99b81081bc9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wpEara96L04+kyro8pH/B1C7wocGL7/laApofKlicxqFIYsTgA7zn/bV0Qs4RFYzAVeYGFaTneNQmj0bQ9/GAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T15:23:43.239088Z","bundle_sha256":"ebb134e465e36f0575997183a046a67ac3219c635253df0b738528912f3e3087"}}