{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:S7IVIMVX3YQ74W6NLHLDF6LHYS","short_pith_number":"pith:S7IVIMVX","canonical_record":{"source":{"id":"2606.20508","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T17:25:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"715966a5c481157fbf1788309e5cf0455f68140463b4a364364b5a74ef1a85f0","abstract_canon_sha256":"c1ecec3d354d5d98046f269d044254d3faf187aab07289bec8da684e3d2bc58f"},"schema_version":"1.0"},"canonical_sha256":"97d15432b7de21fe5bcd59d632f967c4b60f901607716fceaaea8eb36362e5d6","source":{"kind":"arxiv","id":"2606.20508","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.20508","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"arxiv_version","alias_value":"2606.20508v1","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20508","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"pith_short_12","alias_value":"S7IVIMVX3YQ7","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"pith_short_16","alias_value":"S7IVIMVX3YQ74W6N","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"pith_short_8","alias_value":"S7IVIMVX","created_at":"2026-06-19T16:13:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:S7IVIMVX3YQ74W6NLHLDF6LHYS","target":"record","payload":{"canonical_record":{"source":{"id":"2606.20508","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T17:25:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"715966a5c481157fbf1788309e5cf0455f68140463b4a364364b5a74ef1a85f0","abstract_canon_sha256":"c1ecec3d354d5d98046f269d044254d3faf187aab07289bec8da684e3d2bc58f"},"schema_version":"1.0"},"canonical_sha256":"97d15432b7de21fe5bcd59d632f967c4b60f901607716fceaaea8eb36362e5d6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:13:14.027061Z","signature_b64":"G5CJJL35yWzJj4Ie7CwPGZzz6YrFGOlmyRhlWRnhv8PfQGJeDgMjZPrhenHkk37CW59MWULg1w3ogNXXYcZmBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97d15432b7de21fe5bcd59d632f967c4b60f901607716fceaaea8eb36362e5d6","last_reissued_at":"2026-06-19T16:13:14.026718Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:13:14.026718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.20508","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":"2026-06-19T16:13:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GEW7h154GVG6m5vBQkHQn+HpimAgDFCshXphivvzlZ/XvG1gPMiT06vPsk3ocxFNwWudQEEHJHo40A+qOE2hAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T13:09:23.661703Z"},"content_sha256":"07148bd433c5d15079fb1f069d47011c264e004fcccaba65adfaff9009305e4b","schema_version":"1.0","event_id":"sha256:07148bd433c5d15079fb1f069d47011c264e004fcccaba65adfaff9009305e4b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:S7IVIMVX3YQ74W6NLHLDF6LHYS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Mann Patel, Sihui Dai","submitted_at":"2026-06-18T17:25:38Z","abstract_excerpt":"Prior work has shown that in-context demonstrations can jailbreak language models, but it remains unclear how models interpret different types of compliance demonstrations. We study this by mixing benign compliance demonstrations (non-harmful request, helpful response) with harmful compliance demonstrations (harmful request, helpful response) and testing three hypotheses about how demonstration composition drives harmful compliance. Across four models, we find that benign and harmful demonstrations are not interchangeable: benign demonstrations can either reduce or increase harmful compliance "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20508","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/2606.20508/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":"2026-06-19T16:13:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xsLo79z0GsnBKRhpOszhhWGqA3dABt8ZgdN6eWtWJocBLfHFDhwvEbOFFLY94t4Xjfc2DQaesFutKqvGL8YaDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T13:09:23.662079Z"},"content_sha256":"614346eef91c3b901881c80069ee5d8a56a4231943a6f1b22621705d74d83712","schema_version":"1.0","event_id":"sha256:614346eef91c3b901881c80069ee5d8a56a4231943a6f1b22621705d74d83712"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS/bundle.json","state_url":"https://pith.science/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS/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":"2026-07-04T13:09:23Z","links":{"resolver":"https://pith.science/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS","bundle":"https://pith.science/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS/bundle.json","state":"https://pith.science/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/S7IVIMVX3YQ74W6NLHLDF6LHYS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:S7IVIMVX3YQ74W6NLHLDF6LHYS","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":"c1ecec3d354d5d98046f269d044254d3faf187aab07289bec8da684e3d2bc58f","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T17:25:38Z","title_canon_sha256":"715966a5c481157fbf1788309e5cf0455f68140463b4a364364b5a74ef1a85f0"},"schema_version":"1.0","source":{"id":"2606.20508","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.20508","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"arxiv_version","alias_value":"2606.20508v1","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20508","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"pith_short_12","alias_value":"S7IVIMVX3YQ7","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"pith_short_16","alias_value":"S7IVIMVX3YQ74W6N","created_at":"2026-06-19T16:13:14Z"},{"alias_kind":"pith_short_8","alias_value":"S7IVIMVX","created_at":"2026-06-19T16:13:14Z"}],"graph_snapshots":[{"event_id":"sha256:614346eef91c3b901881c80069ee5d8a56a4231943a6f1b22621705d74d83712","target":"graph","created_at":"2026-06-19T16:13:14Z","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/2606.20508/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Prior work has shown that in-context demonstrations can jailbreak language models, but it remains unclear how models interpret different types of compliance demonstrations. We study this by mixing benign compliance demonstrations (non-harmful request, helpful response) with harmful compliance demonstrations (harmful request, helpful response) and testing three hypotheses about how demonstration composition drives harmful compliance. Across four models, we find that benign and harmful demonstrations are not interchangeable: benign demonstrations can either reduce or increase harmful compliance ","authors_text":"Mann Patel, Sihui Dai","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T17:25:38Z","title":"What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20508","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":"sha256:07148bd433c5d15079fb1f069d47011c264e004fcccaba65adfaff9009305e4b","target":"record","created_at":"2026-06-19T16:13:14Z","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":"c1ecec3d354d5d98046f269d044254d3faf187aab07289bec8da684e3d2bc58f","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T17:25:38Z","title_canon_sha256":"715966a5c481157fbf1788309e5cf0455f68140463b4a364364b5a74ef1a85f0"},"schema_version":"1.0","source":{"id":"2606.20508","kind":"arxiv","version":1}},"canonical_sha256":"97d15432b7de21fe5bcd59d632f967c4b60f901607716fceaaea8eb36362e5d6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"97d15432b7de21fe5bcd59d632f967c4b60f901607716fceaaea8eb36362e5d6","first_computed_at":"2026-06-19T16:13:14.026718Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:13:14.026718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"G5CJJL35yWzJj4Ie7CwPGZzz6YrFGOlmyRhlWRnhv8PfQGJeDgMjZPrhenHkk37CW59MWULg1w3ogNXXYcZmBw==","signature_status":"signed_v1","signed_at":"2026-06-19T16:13:14.027061Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.20508","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:07148bd433c5d15079fb1f069d47011c264e004fcccaba65adfaff9009305e4b","sha256:614346eef91c3b901881c80069ee5d8a56a4231943a6f1b22621705d74d83712"],"state_sha256":"ef4db31b58cfa08314d83e86752173e3d764a2f00b4ba476f3275c47e6e26268"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fT5EUEiaHB/gQX5+yGuJgZQz9sYdEmj2/FT3Jj3BCOZLnCSuydb0eavzMcWCrB3T7X6UMQWxyuv2jlglabV7Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T13:09:23.664105Z","bundle_sha256":"ff565626adc2783e07dc40a9753e0457dcd36fbafdaeb8c17a1ab9ee8cb9509c"}}