{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:RUOTT7YSHPJVJMC553ZCHQ4BVM","short_pith_number":"pith:RUOTT7YS","canonical_record":{"source":{"id":"2606.20225","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-18T13:39:59Z","cross_cats_sorted":[],"title_canon_sha256":"51b15f9c640af4fbe46b548bad3a1f02d6895fdd13892598d80892fbc69914e7","abstract_canon_sha256":"bb316df2dd175a6dfa81fe6fcfb3cd9a4de663eba2a132fbed133285c1eb2385"},"schema_version":"1.0"},"canonical_sha256":"8d1d39ff123bd354b05deef223c381ab3316bd094d431c5d2445c96434ffe558","source":{"kind":"arxiv","id":"2606.20225","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.20225","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.20225v1","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20225","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"pith_short_12","alias_value":"RUOTT7YSHPJV","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"pith_short_16","alias_value":"RUOTT7YSHPJVJMC5","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"pith_short_8","alias_value":"RUOTT7YS","created_at":"2026-06-19T16:13:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:RUOTT7YSHPJVJMC553ZCHQ4BVM","target":"record","payload":{"canonical_record":{"source":{"id":"2606.20225","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-18T13:39:59Z","cross_cats_sorted":[],"title_canon_sha256":"51b15f9c640af4fbe46b548bad3a1f02d6895fdd13892598d80892fbc69914e7","abstract_canon_sha256":"bb316df2dd175a6dfa81fe6fcfb3cd9a4de663eba2a132fbed133285c1eb2385"},"schema_version":"1.0"},"canonical_sha256":"8d1d39ff123bd354b05deef223c381ab3316bd094d431c5d2445c96434ffe558","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:13:06.335395Z","signature_b64":"6Ajv2FG8jycAYx0RcGYS5YRfYX1aBsU3j0yDpMzKwFrK4UNnHNpHTrBBPGenXwtbljRO9cmeo5XGlRWcjaLFCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d1d39ff123bd354b05deef223c381ab3316bd094d431c5d2445c96434ffe558","last_reissued_at":"2026-06-19T16:13:06.334970Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:13:06.334970Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.20225","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":"2026-06-19T16:13:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OWcDsTFvygw7x2daY3hVKtLFDXfAd5ouwTxYdcLjl82hnbaudswRuoOr83FGebhXZLbiZDLmfjnYUd2jToiYBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T06:48:37.108186Z"},"content_sha256":"ec1a25f67cd7820d7a830e2b307d78fc1a7748004dd867c2832cc53e5bdaab55","schema_version":"1.0","event_id":"sha256:ec1a25f67cd7820d7a830e2b307d78fc1a7748004dd867c2832cc53e5bdaab55"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:RUOTT7YSHPJVJMC553ZCHQ4BVM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Actionable Activation Directions for Detecting and Mitigating Emergent Misalignment Across Language Model Families","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abdul Rafay Syed","submitted_at":"2026-06-18T13:39:59Z","abstract_excerpt":"Fine-tuning language models on insecure code induces emergent misalignment with poorly understood internal structure. We investigate whether this misalignment corresponds to a causally actionable activation-space direction shared across architectures. Across four instruction-tuned model families (Qwen2.5-1.5B, Gemma-2-2B, Llama-3.2-1B, Ministral-3-3B) finetuned identically, a difference-in-means direction achieves 99.6% separation of aligned and misaligned activations at each model's final layer. Causal steering by subtracting this direction reduces code spillover by 21-51 points, while a secu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20225","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/2606.20225/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":"2026-06-19T16:13:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B5nIdh7PG6idfmtIEAST8XthcCT2CY84WRE2kRAWH0cg+7pUYZrOu8tUg5HOfXYj9X3ULo3Cq3GkInQY+FqgAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T06:48:37.108572Z"},"content_sha256":"365f14b28865701a1e81ed9f16a7d51f72ceb3d9b8578f6a3160c3b3be2aeffc","schema_version":"1.0","event_id":"sha256:365f14b28865701a1e81ed9f16a7d51f72ceb3d9b8578f6a3160c3b3be2aeffc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM/bundle.json","state_url":"https://pith.science/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM/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":"2026-06-28T06:48:37Z","links":{"resolver":"https://pith.science/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM","bundle":"https://pith.science/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM/bundle.json","state":"https://pith.science/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RUOTT7YSHPJVJMC553ZCHQ4BVM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RUOTT7YSHPJVJMC553ZCHQ4BVM","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":"bb316df2dd175a6dfa81fe6fcfb3cd9a4de663eba2a132fbed133285c1eb2385","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-18T13:39:59Z","title_canon_sha256":"51b15f9c640af4fbe46b548bad3a1f02d6895fdd13892598d80892fbc69914e7"},"schema_version":"1.0","source":{"id":"2606.20225","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.20225","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.20225v1","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20225","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"pith_short_12","alias_value":"RUOTT7YSHPJV","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"pith_short_16","alias_value":"RUOTT7YSHPJVJMC5","created_at":"2026-06-19T16:13:06Z"},{"alias_kind":"pith_short_8","alias_value":"RUOTT7YS","created_at":"2026-06-19T16:13:06Z"}],"graph_snapshots":[{"event_id":"sha256:365f14b28865701a1e81ed9f16a7d51f72ceb3d9b8578f6a3160c3b3be2aeffc","target":"graph","created_at":"2026-06-19T16:13:06Z","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/2606.20225/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning language models on insecure code induces emergent misalignment with poorly understood internal structure. We investigate whether this misalignment corresponds to a causally actionable activation-space direction shared across architectures. Across four instruction-tuned model families (Qwen2.5-1.5B, Gemma-2-2B, Llama-3.2-1B, Ministral-3-3B) finetuned identically, a difference-in-means direction achieves 99.6% separation of aligned and misaligned activations at each model's final layer. Causal steering by subtracting this direction reduces code spillover by 21-51 points, while a secu","authors_text":"Abdul Rafay Syed","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-18T13:39:59Z","title":"Actionable Activation Directions for Detecting and Mitigating Emergent Misalignment Across Language Model Families"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20225","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:ec1a25f67cd7820d7a830e2b307d78fc1a7748004dd867c2832cc53e5bdaab55","target":"record","created_at":"2026-06-19T16:13:06Z","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":"bb316df2dd175a6dfa81fe6fcfb3cd9a4de663eba2a132fbed133285c1eb2385","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-18T13:39:59Z","title_canon_sha256":"51b15f9c640af4fbe46b548bad3a1f02d6895fdd13892598d80892fbc69914e7"},"schema_version":"1.0","source":{"id":"2606.20225","kind":"arxiv","version":1}},"canonical_sha256":"8d1d39ff123bd354b05deef223c381ab3316bd094d431c5d2445c96434ffe558","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8d1d39ff123bd354b05deef223c381ab3316bd094d431c5d2445c96434ffe558","first_computed_at":"2026-06-19T16:13:06.334970Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:13:06.334970Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6Ajv2FG8jycAYx0RcGYS5YRfYX1aBsU3j0yDpMzKwFrK4UNnHNpHTrBBPGenXwtbljRO9cmeo5XGlRWcjaLFCg==","signature_status":"signed_v1","signed_at":"2026-06-19T16:13:06.335395Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.20225","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ec1a25f67cd7820d7a830e2b307d78fc1a7748004dd867c2832cc53e5bdaab55","sha256:365f14b28865701a1e81ed9f16a7d51f72ceb3d9b8578f6a3160c3b3be2aeffc"],"state_sha256":"0ac7b8a117b2d009787f3d1a81fee95c8cdc9fa422e21f4a3e0b820f699cdc17"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UqScpAKTCQhbv5YATPEQu5rptVZVDOh7B5KMjxFRrV2FizdzysbRFpknLM4wy/HqNkqpkas2P9+SiNeghlo9Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T06:48:37.110558Z","bundle_sha256":"188a225fda017753f32ceed62877d7983a5f7c9d3b655316f75dcea922705e90"}}