{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TFW2CZVWIT22CPHGV37PMWWS7A","short_pith_number":"pith:TFW2CZVW","canonical_record":{"source":{"id":"2606.21077","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-19T03:55:08Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a3c2c2b891d1ff38058e4c79e5dfb4b7cdb2dfc617af89f60844bc0c1def6dbd","abstract_canon_sha256":"525feceee5617cf75f050d2e04c137b9df000f4c023058602c9c7faa77924799"},"schema_version":"1.0"},"canonical_sha256":"996da166b644f5a13ce6aefef65ad2f820d83584089eeebf9edb2ac20fb69133","source":{"kind":"arxiv","id":"2606.21077","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.21077","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"arxiv_version","alias_value":"2606.21077v1","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21077","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"pith_short_12","alias_value":"TFW2CZVWIT22","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"pith_short_16","alias_value":"TFW2CZVWIT22CPHG","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"pith_short_8","alias_value":"TFW2CZVW","created_at":"2026-06-23T01:12:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TFW2CZVWIT22CPHGV37PMWWS7A","target":"record","payload":{"canonical_record":{"source":{"id":"2606.21077","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-19T03:55:08Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a3c2c2b891d1ff38058e4c79e5dfb4b7cdb2dfc617af89f60844bc0c1def6dbd","abstract_canon_sha256":"525feceee5617cf75f050d2e04c137b9df000f4c023058602c9c7faa77924799"},"schema_version":"1.0"},"canonical_sha256":"996da166b644f5a13ce6aefef65ad2f820d83584089eeebf9edb2ac20fb69133","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:29.082068Z","signature_b64":"mnnC0Z0cTLz6CDJba8PyjqsjeFclF4xwy1vqkdFjnczQhnOrgMrh7HVxC07AZ4gYqaZCBC7n6lrOneIJy8whDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"996da166b644f5a13ce6aefef65ad2f820d83584089eeebf9edb2ac20fb69133","last_reissued_at":"2026-06-23T01:12:29.081565Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:29.081565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.21077","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":"2026-06-23T01:12:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"azMxnwX8oQHGx/dMGwHtzHYRJOlfUuMias+DJolZIF+kzVX77jtVsQ3KkLQqq3b1XsPdjK1gdGGLbvxzOMYwBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-02T22:50:50.838830Z"},"content_sha256":"89119b0e3d43dae5e7bcdf566899934a13177e8261386fa1d5b637f1c3fd4214","schema_version":"1.0","event_id":"sha256:89119b0e3d43dae5e7bcdf566899934a13177e8261386fa1d5b637f1c3fd4214"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TFW2CZVWIT22CPHGV37PMWWS7A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CR","authors_text":"Fang Yu, Hsin-Ling Hsu, Jerry Wang, Nai-Chia Chen, Yi-Cheng Lai","submitted_at":"2026-06-19T03:55:08Z","abstract_excerpt":"Production LLMs increasingly rely on toxicity-based moderation filters as a primary defense, assuming that harmful intent correlates with toxic surface wording. We show this assumption is fundamentally brittle: surface toxicity and adversarial intent can be decoupled by replacing as few as five tokens. We present OTTER (Obfuscated Toxicity-Evading Token Evolution for Rewriting), a black-box red-teaming framework requiring only standard API access, directly targeting the practical constraints of industry security audits. Evaluated on 457 AdvBench prompts across four GPT models, OTTER raises ave"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21077","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/2606.21077/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":"2026-06-23T01:12:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KtsYvn5MGBUR+jmyptCY9Lptm4BfPqyKjqnGMgTnlg86h4HCt5dWt4YV8AAhrqg06jYrOjkvPM8Usz5K1hYdCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-02T22:50:50.839200Z"},"content_sha256":"2556f98a6134da387132da53101b78db6b1b751e854d96b070e3bd260ab57a87","schema_version":"1.0","event_id":"sha256:2556f98a6134da387132da53101b78db6b1b751e854d96b070e3bd260ab57a87"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TFW2CZVWIT22CPHGV37PMWWS7A/bundle.json","state_url":"https://pith.science/pith/TFW2CZVWIT22CPHGV37PMWWS7A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TFW2CZVWIT22CPHGV37PMWWS7A/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":"2026-07-02T22:50:50Z","links":{"resolver":"https://pith.science/pith/TFW2CZVWIT22CPHGV37PMWWS7A","bundle":"https://pith.science/pith/TFW2CZVWIT22CPHGV37PMWWS7A/bundle.json","state":"https://pith.science/pith/TFW2CZVWIT22CPHGV37PMWWS7A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TFW2CZVWIT22CPHGV37PMWWS7A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TFW2CZVWIT22CPHGV37PMWWS7A","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":"525feceee5617cf75f050d2e04c137b9df000f4c023058602c9c7faa77924799","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-19T03:55:08Z","title_canon_sha256":"a3c2c2b891d1ff38058e4c79e5dfb4b7cdb2dfc617af89f60844bc0c1def6dbd"},"schema_version":"1.0","source":{"id":"2606.21077","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.21077","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"arxiv_version","alias_value":"2606.21077v1","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21077","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"pith_short_12","alias_value":"TFW2CZVWIT22","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"pith_short_16","alias_value":"TFW2CZVWIT22CPHG","created_at":"2026-06-23T01:12:29Z"},{"alias_kind":"pith_short_8","alias_value":"TFW2CZVW","created_at":"2026-06-23T01:12:29Z"}],"graph_snapshots":[{"event_id":"sha256:2556f98a6134da387132da53101b78db6b1b751e854d96b070e3bd260ab57a87","target":"graph","created_at":"2026-06-23T01:12:29Z","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/2606.21077/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Production LLMs increasingly rely on toxicity-based moderation filters as a primary defense, assuming that harmful intent correlates with toxic surface wording. We show this assumption is fundamentally brittle: surface toxicity and adversarial intent can be decoupled by replacing as few as five tokens. We present OTTER (Obfuscated Toxicity-Evading Token Evolution for Rewriting), a black-box red-teaming framework requiring only standard API access, directly targeting the practical constraints of industry security audits. Evaluated on 457 AdvBench prompts across four GPT models, OTTER raises ave","authors_text":"Fang Yu, Hsin-Ling Hsu, Jerry Wang, Nai-Chia Chen, Yi-Cheng Lai","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-19T03:55:08Z","title":"OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21077","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":"sha256:89119b0e3d43dae5e7bcdf566899934a13177e8261386fa1d5b637f1c3fd4214","target":"record","created_at":"2026-06-23T01:12:29Z","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":"525feceee5617cf75f050d2e04c137b9df000f4c023058602c9c7faa77924799","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-19T03:55:08Z","title_canon_sha256":"a3c2c2b891d1ff38058e4c79e5dfb4b7cdb2dfc617af89f60844bc0c1def6dbd"},"schema_version":"1.0","source":{"id":"2606.21077","kind":"arxiv","version":1}},"canonical_sha256":"996da166b644f5a13ce6aefef65ad2f820d83584089eeebf9edb2ac20fb69133","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"996da166b644f5a13ce6aefef65ad2f820d83584089eeebf9edb2ac20fb69133","first_computed_at":"2026-06-23T01:12:29.081565Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T01:12:29.081565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mnnC0Z0cTLz6CDJba8PyjqsjeFclF4xwy1vqkdFjnczQhnOrgMrh7HVxC07AZ4gYqaZCBC7n6lrOneIJy8whDA==","signature_status":"signed_v1","signed_at":"2026-06-23T01:12:29.082068Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.21077","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:89119b0e3d43dae5e7bcdf566899934a13177e8261386fa1d5b637f1c3fd4214","sha256:2556f98a6134da387132da53101b78db6b1b751e854d96b070e3bd260ab57a87"],"state_sha256":"1b167f37b3a8b740dfb17b949fe1377247bb77614e65fa9bae43b99294e445ba"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IRQ1hhenBTaBYvEdfivTwzUprq/yO7vxve6wOsQlWd9DLJp5o1ccL+s48CvLDsW/PKwQKyblXhRqBKdVz/A4AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-02T22:50:50.841383Z","bundle_sha256":"3b5931e4fd17b23404e9a358891f0bcfb7fb45e90f6686160babddfd9c2bedda"}}