{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:Z7SBT5PEZRZOZDA2LHWWO6VSQO","short_pith_number":"pith:Z7SBT5PE","canonical_record":{"source":{"id":"2501.08246","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T16:32:01Z","cross_cats_sorted":[],"title_canon_sha256":"b1a1228364a21c288e3ace6e9772efb5a169af7c792c6d8b58ee5042cb12e599","abstract_canon_sha256":"ca5e7aa11a1601a00a46167005bf81b07e0fa78b225816c9eece5c6a7a50ff8f"},"schema_version":"1.0"},"canonical_sha256":"cfe419f5e4cc72ec8c1a59ed677ab283be31288f3a56b9e91cfa63106d66129f","source":{"kind":"arxiv","id":"2501.08246","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.08246","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"arxiv_version","alias_value":"2501.08246v1","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08246","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"pith_short_12","alias_value":"Z7SBT5PEZRZO","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"pith_short_16","alias_value":"Z7SBT5PEZRZOZDA2","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"pith_short_8","alias_value":"Z7SBT5PE","created_at":"2026-07-05T10:00:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:Z7SBT5PEZRZOZDA2LHWWO6VSQO","target":"record","payload":{"canonical_record":{"source":{"id":"2501.08246","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T16:32:01Z","cross_cats_sorted":[],"title_canon_sha256":"b1a1228364a21c288e3ace6e9772efb5a169af7c792c6d8b58ee5042cb12e599","abstract_canon_sha256":"ca5e7aa11a1601a00a46167005bf81b07e0fa78b225816c9eece5c6a7a50ff8f"},"schema_version":"1.0"},"canonical_sha256":"cfe419f5e4cc72ec8c1a59ed677ab283be31288f3a56b9e91cfa63106d66129f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:00:58.449900Z","signature_b64":"juOiWOkIxpGfgLMkO7QOsnVuzy1Ll0Kdz7nH5E3c/9lU6dlyI/4qw6uqSj0kUcWMGifR+UB/2CjtdEqhKdHRCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cfe419f5e4cc72ec8c1a59ed677ab283be31288f3a56b9e91cfa63106d66129f","last_reissued_at":"2026-07-05T10:00:58.449483Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:00:58.449483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.08246","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":"2026-07-05T10:00:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PKVwOi0TcuglBCxugl5BPqf0OiChEb63+8IPT1ZrNElvdN1lcNzqz1biQUvsNLH67CJoilUMQqgqMFtWP5DhAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T19:49:30.640839Z"},"content_sha256":"54213e391621299fb5f06825797de625724fab36387eacc0b91e0c49a1f49961","schema_version":"1.0","event_id":"sha256:54213e391621299fb5f06825797de625724fab36387eacc0b91e0c49a1f49961"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:Z7SBT5PEZRZOZDA2LHWWO6VSQO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Text-Diffusion Red-Teaming of Large Language Models: Unveiling Harmful Behaviors with Proximity Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Adish Singla, Goran Radanovi\\'c, Jonathan N\\\"other","submitted_at":"2025-01-14T16:32:01Z","abstract_excerpt":"Recent work has proposed automated red-teaming methods for testing the vulnerabilities of a given target large language model (LLM). These methods use red-teaming LLMs to uncover inputs that induce harmful behavior in a target LLM. In this paper, we study red-teaming strategies that enable a targeted security assessment. We propose an optimization framework for red-teaming with proximity constraints, where the discovered prompts must be similar to reference prompts from a given dataset. This dataset serves as a template for the discovered prompts, anchoring the search for test-cases to specifi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08246","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/2501.08246/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":"2026-07-05T10:00:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aPujbRhLbIOB5RWxFA+34TF5T2pkyQae1BHE3Pvf6TUaZBMqZAL9HcF9g/Fmh8Vcx8zXZV50sapDqd4WOn5UCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T19:49:30.641427Z"},"content_sha256":"f55e2039f98f75f540d222233cf3f3095002ebacc133098528dce6cd2fbe49d4","schema_version":"1.0","event_id":"sha256:f55e2039f98f75f540d222233cf3f3095002ebacc133098528dce6cd2fbe49d4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO/bundle.json","state_url":"https://pith.science/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO/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":"2026-08-12T19:49:30Z","links":{"resolver":"https://pith.science/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO","bundle":"https://pith.science/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO/bundle.json","state":"https://pith.science/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Z7SBT5PEZRZOZDA2LHWWO6VSQO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Z7SBT5PEZRZOZDA2LHWWO6VSQO","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":"ca5e7aa11a1601a00a46167005bf81b07e0fa78b225816c9eece5c6a7a50ff8f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T16:32:01Z","title_canon_sha256":"b1a1228364a21c288e3ace6e9772efb5a169af7c792c6d8b58ee5042cb12e599"},"schema_version":"1.0","source":{"id":"2501.08246","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.08246","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"arxiv_version","alias_value":"2501.08246v1","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08246","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"pith_short_12","alias_value":"Z7SBT5PEZRZO","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"pith_short_16","alias_value":"Z7SBT5PEZRZOZDA2","created_at":"2026-07-05T10:00:58Z"},{"alias_kind":"pith_short_8","alias_value":"Z7SBT5PE","created_at":"2026-07-05T10:00:58Z"}],"graph_snapshots":[{"event_id":"sha256:f55e2039f98f75f540d222233cf3f3095002ebacc133098528dce6cd2fbe49d4","target":"graph","created_at":"2026-07-05T10:00:58Z","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/2501.08246/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work has proposed automated red-teaming methods for testing the vulnerabilities of a given target large language model (LLM). These methods use red-teaming LLMs to uncover inputs that induce harmful behavior in a target LLM. In this paper, we study red-teaming strategies that enable a targeted security assessment. We propose an optimization framework for red-teaming with proximity constraints, where the discovered prompts must be similar to reference prompts from a given dataset. This dataset serves as a template for the discovered prompts, anchoring the search for test-cases to specifi","authors_text":"Adish Singla, Goran Radanovi\\'c, Jonathan N\\\"other","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T16:32:01Z","title":"Text-Diffusion Red-Teaming of Large Language Models: Unveiling Harmful Behaviors with Proximity Constraints"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08246","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":"sha256:54213e391621299fb5f06825797de625724fab36387eacc0b91e0c49a1f49961","target":"record","created_at":"2026-07-05T10:00:58Z","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":"ca5e7aa11a1601a00a46167005bf81b07e0fa78b225816c9eece5c6a7a50ff8f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T16:32:01Z","title_canon_sha256":"b1a1228364a21c288e3ace6e9772efb5a169af7c792c6d8b58ee5042cb12e599"},"schema_version":"1.0","source":{"id":"2501.08246","kind":"arxiv","version":1}},"canonical_sha256":"cfe419f5e4cc72ec8c1a59ed677ab283be31288f3a56b9e91cfa63106d66129f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cfe419f5e4cc72ec8c1a59ed677ab283be31288f3a56b9e91cfa63106d66129f","first_computed_at":"2026-07-05T10:00:58.449483Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:00:58.449483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"juOiWOkIxpGfgLMkO7QOsnVuzy1Ll0Kdz7nH5E3c/9lU6dlyI/4qw6uqSj0kUcWMGifR+UB/2CjtdEqhKdHRCw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:00:58.449900Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.08246","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54213e391621299fb5f06825797de625724fab36387eacc0b91e0c49a1f49961","sha256:f55e2039f98f75f540d222233cf3f3095002ebacc133098528dce6cd2fbe49d4"],"state_sha256":"7d6c15138fb3df449b34a12261e54d2a732cd609fb2164a7252f70774af61fb5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mlu2zWOq2g+2qdPP+iqoWYT9U9zoP0uHM3/YJmUD0TqelY8oiRxr4Wel7XKhKKi8AFLtHsUZbr34DEsKBm+DDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T19:49:30.645953Z","bundle_sha256":"5debb16f691b3cd332eec0573bfb02cc649849702c03133a3f4236008636609a"}}