{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:R6OQRLJGZGZNAGNAPWBEQIQM5H","short_pith_number":"pith:R6OQRLJG","canonical_record":{"source":{"id":"2506.00085","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T04:54:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9b09695c28f48d3c2b71b19f0468b6519bf40da55184602832d47406bea6fe2e","abstract_canon_sha256":"1aab16ee514b4e86d17dee503f5a7a52b8fcf2b4b6c11db675611a90565b483d"},"schema_version":"1.0"},"canonical_sha256":"8f9d08ad26c9b2d019a07d8248220ce9faaa48881cfde9b7cdfa0e545c033814","source":{"kind":"arxiv","id":"2506.00085","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.00085","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"arxiv_version","alias_value":"2506.00085v1","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00085","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"pith_short_12","alias_value":"R6OQRLJGZGZN","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"pith_short_16","alias_value":"R6OQRLJGZGZNAGNA","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"pith_short_8","alias_value":"R6OQRLJG","created_at":"2026-07-05T11:13:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:R6OQRLJGZGZNAGNAPWBEQIQM5H","target":"record","payload":{"canonical_record":{"source":{"id":"2506.00085","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T04:54:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9b09695c28f48d3c2b71b19f0468b6519bf40da55184602832d47406bea6fe2e","abstract_canon_sha256":"1aab16ee514b4e86d17dee503f5a7a52b8fcf2b4b6c11db675611a90565b483d"},"schema_version":"1.0"},"canonical_sha256":"8f9d08ad26c9b2d019a07d8248220ce9faaa48881cfde9b7cdfa0e545c033814","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:28.796344Z","signature_b64":"x4ArDPupGWGcjRcoegsZg/VuIfPgC3m0GpVB/tJ5Ud9upPzHeQlbLD0NmhdH+MXRzt+8BIwL+v0uuw/6WscqAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f9d08ad26c9b2d019a07d8248220ce9faaa48881cfde9b7cdfa0e545c033814","last_reissued_at":"2026-07-05T11:13:28.795883Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:28.795883Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.00085","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":"2026-07-05T11:13:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/eIDDDT27O8kedXlE/yvWL9XhJgiRl6qug922bHTkGLkO+wZV5nAYWKLHPO4Uw/AuWQzAUqZCGn1yIrsNvxjAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T10:33:09.952505Z"},"content_sha256":"bd1397e91377c0c040ca572ab4d8e60df52498ea0dd05c084c2e772450f0dc61","schema_version":"1.0","event_id":"sha256:bd1397e91377c0c040ca572ab4d8e60df52498ea0dd05c084c2e772450f0dc61"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:R6OQRLJGZGZNAGNAPWBEQIQM5H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"COSMIC: Generalized Refusal Direction Identification in LLM Activations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenguang Wang, Dawn Song, Nicholas Crispino, Sam Pan, Vincent Siu, Yang Liu, Zhun Wang, Zihao Yu","submitted_at":"2025-05-30T04:54:18Z","abstract_excerpt":"Large Language Models (LLMs) encode behaviors such as refusal within their activation space, yet identifying these behaviors remains a significant challenge. Existing methods often rely on predefined refusal templates detectable in output tokens or require manual analysis. We introduce \\textbf{COSMIC} (Cosine Similarity Metrics for Inversion of Concepts), an automated framework for direction selection that identifies viable steering directions and target layers using cosine similarity - entirely independent of model outputs. COSMIC achieves steering performance comparable to prior methods with"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00085","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/2506.00085/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":"2026-07-05T11:13:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A1tr0iGPvNQ0voCRJjoxxO5v8k4Xr95FMCojZTy6e8A7lUUyvGd7u/zNaGrujOt1VQv1sYlFw8V+SIVtXbQODw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T10:33:09.953432Z"},"content_sha256":"e9c92c0d5b2018f5ffa7a1fa28d4a2226e19d4fe4822276e203f0e934296c3ff","schema_version":"1.0","event_id":"sha256:e9c92c0d5b2018f5ffa7a1fa28d4a2226e19d4fe4822276e203f0e934296c3ff"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H/bundle.json","state_url":"https://pith.science/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H/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":"2026-08-18T10:33:09Z","links":{"resolver":"https://pith.science/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H","bundle":"https://pith.science/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H/bundle.json","state":"https://pith.science/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R6OQRLJGZGZNAGNAPWBEQIQM5H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:R6OQRLJGZGZNAGNAPWBEQIQM5H","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":"1aab16ee514b4e86d17dee503f5a7a52b8fcf2b4b6c11db675611a90565b483d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T04:54:18Z","title_canon_sha256":"9b09695c28f48d3c2b71b19f0468b6519bf40da55184602832d47406bea6fe2e"},"schema_version":"1.0","source":{"id":"2506.00085","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.00085","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"arxiv_version","alias_value":"2506.00085v1","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00085","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"pith_short_12","alias_value":"R6OQRLJGZGZN","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"pith_short_16","alias_value":"R6OQRLJGZGZNAGNA","created_at":"2026-07-05T11:13:28Z"},{"alias_kind":"pith_short_8","alias_value":"R6OQRLJG","created_at":"2026-07-05T11:13:28Z"}],"graph_snapshots":[{"event_id":"sha256:e9c92c0d5b2018f5ffa7a1fa28d4a2226e19d4fe4822276e203f0e934296c3ff","target":"graph","created_at":"2026-07-05T11:13:28Z","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/2506.00085/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) encode behaviors such as refusal within their activation space, yet identifying these behaviors remains a significant challenge. Existing methods often rely on predefined refusal templates detectable in output tokens or require manual analysis. We introduce \\textbf{COSMIC} (Cosine Similarity Metrics for Inversion of Concepts), an automated framework for direction selection that identifies viable steering directions and target layers using cosine similarity - entirely independent of model outputs. COSMIC achieves steering performance comparable to prior methods with","authors_text":"Chenguang Wang, Dawn Song, Nicholas Crispino, Sam Pan, Vincent Siu, Yang Liu, Zhun Wang, Zihao Yu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T04:54:18Z","title":"COSMIC: Generalized Refusal Direction Identification in LLM Activations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00085","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:bd1397e91377c0c040ca572ab4d8e60df52498ea0dd05c084c2e772450f0dc61","target":"record","created_at":"2026-07-05T11:13:28Z","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":"1aab16ee514b4e86d17dee503f5a7a52b8fcf2b4b6c11db675611a90565b483d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T04:54:18Z","title_canon_sha256":"9b09695c28f48d3c2b71b19f0468b6519bf40da55184602832d47406bea6fe2e"},"schema_version":"1.0","source":{"id":"2506.00085","kind":"arxiv","version":1}},"canonical_sha256":"8f9d08ad26c9b2d019a07d8248220ce9faaa48881cfde9b7cdfa0e545c033814","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8f9d08ad26c9b2d019a07d8248220ce9faaa48881cfde9b7cdfa0e545c033814","first_computed_at":"2026-07-05T11:13:28.795883Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:28.795883Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"x4ArDPupGWGcjRcoegsZg/VuIfPgC3m0GpVB/tJ5Ud9upPzHeQlbLD0NmhdH+MXRzt+8BIwL+v0uuw/6WscqAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:28.796344Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.00085","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bd1397e91377c0c040ca572ab4d8e60df52498ea0dd05c084c2e772450f0dc61","sha256:e9c92c0d5b2018f5ffa7a1fa28d4a2226e19d4fe4822276e203f0e934296c3ff"],"state_sha256":"32a5231c7ee7045397ea1225f8abc98a095f04369a5223c0db44e7ed10982b1e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iAJ4xgibhgLdCmjdR1uN+5wOGSsqqFNwTiEsN7dbCX6OSKthOzkKN8JK22bRcPSkllklvt4+z45UG6Ok4k0EDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T10:33:09.963923Z","bundle_sha256":"5f8d385e9f521a7736d4303ae98e14f54057b550a12380d3a212f655442b87d8"}}