{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:AV7W2TNT5R4TWZS6VJJLW6HY4D","short_pith_number":"pith:AV7W2TNT","canonical_record":{"source":{"id":"2406.18510","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T17:31:22Z","cross_cats_sorted":[],"title_canon_sha256":"3804d8e992a5c33f2a671ee1230e80209ff10e78054c08cd707046714c32ec77","abstract_canon_sha256":"2852bfbdb876798d549e42e386f090a21c5cdf4c7137c04e574236d213a7ca50"},"schema_version":"1.0"},"canonical_sha256":"057f6d4db3ec793b665eaa52bb78f8e0eadb51e07d270a9d03042b8dc3b48407","source":{"kind":"arxiv","id":"2406.18510","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18510","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18510v1","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18510","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"pith_short_12","alias_value":"AV7W2TNT5R4T","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"pith_short_16","alias_value":"AV7W2TNT5R4TWZS6","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"pith_short_8","alias_value":"AV7W2TNT","created_at":"2026-07-05T08:37:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:AV7W2TNT5R4TWZS6VJJLW6HY4D","target":"record","payload":{"canonical_record":{"source":{"id":"2406.18510","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T17:31:22Z","cross_cats_sorted":[],"title_canon_sha256":"3804d8e992a5c33f2a671ee1230e80209ff10e78054c08cd707046714c32ec77","abstract_canon_sha256":"2852bfbdb876798d549e42e386f090a21c5cdf4c7137c04e574236d213a7ca50"},"schema_version":"1.0"},"canonical_sha256":"057f6d4db3ec793b665eaa52bb78f8e0eadb51e07d270a9d03042b8dc3b48407","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:37:10.308586Z","signature_b64":"ovr7db+XgPr53MqSI5s05EjjfeEbIbaUtcT1B0mYliUlfhPfIgr+E6GH6EkAJoLYElsG+4hzXPWpUNwb3y9eDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"057f6d4db3ec793b665eaa52bb78f8e0eadb51e07d270a9d03042b8dc3b48407","last_reissued_at":"2026-07-05T08:37:10.308113Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:37:10.308113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.18510","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":"2026-07-05T08:37:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lBwcINfowDgEYzZzwvN2GphDPzM9OwmaeAunxK8AQwj7jmNwn9+AG+rLDSkMx6qWJsOgQ4OhFMk/VtSMaYScAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T15:33:45.825913Z"},"content_sha256":"35e4911f12effbda4421ee645c1066ea5a93d85cbc5c1988be79f922297c6881","schema_version":"1.0","event_id":"sha256:35e4911f12effbda4421ee645c1066ea5a93d85cbc5c1988be79f922297c6881"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:AV7W2TNT5R4TWZS6VJJLW6HY4D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Allyson Ettinger, Faeze Brahman, Kavel Rao, Liwei Jiang, Maarten Sap, Niloofar Mireshghallah, Nouha Dziri, Sachin Kumar, Seungju Han, Ximing Lu, Yejin Choi","submitted_at":"2024-06-26T17:31:22Z","abstract_excerpt":"We introduce WildTeaming, an automatic LLM safety red-teaming framework that mines in-the-wild user-chatbot interactions to discover 5.7K unique clusters of novel jailbreak tactics, and then composes multiple tactics for systematic exploration of novel jailbreaks. Compared to prior work that performed red-teaming via recruited human workers, gradient-based optimization, or iterative revision with LLMs, our work investigates jailbreaks from chatbot users who were not specifically instructed to break the system. WildTeaming reveals previously unidentified vulnerabilities of frontier LLMs, result"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18510","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/2406.18510/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":"2026-07-05T08:37:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ri305x+IQkSvlin3hz49F9sXSuLAatJh6JvK8I8OjaFjQsLldbHYwKU+pyqj8dx/6fBXDl9t7KHh21gvzudQCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T15:33:45.826415Z"},"content_sha256":"ec14edf596cfdb5f2b68f8dec47a636e4d68b479725e2e082e045e2e113aab09","schema_version":"1.0","event_id":"sha256:ec14edf596cfdb5f2b68f8dec47a636e4d68b479725e2e082e045e2e113aab09"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D/bundle.json","state_url":"https://pith.science/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D/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":"2026-08-03T15:33:45Z","links":{"resolver":"https://pith.science/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D","bundle":"https://pith.science/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D/bundle.json","state":"https://pith.science/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AV7W2TNT5R4TWZS6VJJLW6HY4D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AV7W2TNT5R4TWZS6VJJLW6HY4D","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":"2852bfbdb876798d549e42e386f090a21c5cdf4c7137c04e574236d213a7ca50","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T17:31:22Z","title_canon_sha256":"3804d8e992a5c33f2a671ee1230e80209ff10e78054c08cd707046714c32ec77"},"schema_version":"1.0","source":{"id":"2406.18510","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18510","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18510v1","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18510","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"pith_short_12","alias_value":"AV7W2TNT5R4T","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"pith_short_16","alias_value":"AV7W2TNT5R4TWZS6","created_at":"2026-07-05T08:37:10Z"},{"alias_kind":"pith_short_8","alias_value":"AV7W2TNT","created_at":"2026-07-05T08:37:10Z"}],"graph_snapshots":[{"event_id":"sha256:ec14edf596cfdb5f2b68f8dec47a636e4d68b479725e2e082e045e2e113aab09","target":"graph","created_at":"2026-07-05T08:37:10Z","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/2406.18510/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce WildTeaming, an automatic LLM safety red-teaming framework that mines in-the-wild user-chatbot interactions to discover 5.7K unique clusters of novel jailbreak tactics, and then composes multiple tactics for systematic exploration of novel jailbreaks. Compared to prior work that performed red-teaming via recruited human workers, gradient-based optimization, or iterative revision with LLMs, our work investigates jailbreaks from chatbot users who were not specifically instructed to break the system. WildTeaming reveals previously unidentified vulnerabilities of frontier LLMs, result","authors_text":"Allyson Ettinger, Faeze Brahman, Kavel Rao, Liwei Jiang, Maarten Sap, Niloofar Mireshghallah, Nouha Dziri, Sachin Kumar, Seungju Han, Ximing Lu, Yejin Choi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T17:31:22Z","title":"WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18510","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":"sha256:35e4911f12effbda4421ee645c1066ea5a93d85cbc5c1988be79f922297c6881","target":"record","created_at":"2026-07-05T08:37:10Z","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":"2852bfbdb876798d549e42e386f090a21c5cdf4c7137c04e574236d213a7ca50","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T17:31:22Z","title_canon_sha256":"3804d8e992a5c33f2a671ee1230e80209ff10e78054c08cd707046714c32ec77"},"schema_version":"1.0","source":{"id":"2406.18510","kind":"arxiv","version":1}},"canonical_sha256":"057f6d4db3ec793b665eaa52bb78f8e0eadb51e07d270a9d03042b8dc3b48407","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"057f6d4db3ec793b665eaa52bb78f8e0eadb51e07d270a9d03042b8dc3b48407","first_computed_at":"2026-07-05T08:37:10.308113Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:37:10.308113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ovr7db+XgPr53MqSI5s05EjjfeEbIbaUtcT1B0mYliUlfhPfIgr+E6GH6EkAJoLYElsG+4hzXPWpUNwb3y9eDA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:37:10.308586Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.18510","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:35e4911f12effbda4421ee645c1066ea5a93d85cbc5c1988be79f922297c6881","sha256:ec14edf596cfdb5f2b68f8dec47a636e4d68b479725e2e082e045e2e113aab09"],"state_sha256":"c570c9a2b5f2632b1cd5e3bd5b245c26aa0805fd5a5ae2d5383c3403dcad183d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9F9Pgq8esEEhnRr53TPArXKSEIuepP93B2l/pOnje1Fpi0IF6bvF1CeIvrK+Stbp/sVvatQZPiQuUreR/u4eBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T15:33:45.830335Z","bundle_sha256":"2009194c64028534b10dc29467cdec6c578a9bb1e7b0b7da78162c53a72bfaf9"}}