{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Y6USGH5PMTN5GVNMXQAEMHH3BE","short_pith_number":"pith:Y6USGH5P","canonical_record":{"source":{"id":"2409.05907","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T15:47:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"1c40c3cde09063ae75c93e0d8b66503f2cc931085f26afc8dcab555ef664294a","abstract_canon_sha256":"e5685333ba0c99c8f10662f44ffd475738d8ad92b7c43a6d6080b31a7ae1f16b"},"schema_version":"1.0"},"canonical_sha256":"c7a9231faf64dbd355acbc00461cfb091d0e65bac6c9a66865d82e52a192d7eb","source":{"kind":"arxiv","id":"2409.05907","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.05907","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"arxiv_version","alias_value":"2409.05907v3","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.05907","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"pith_short_12","alias_value":"Y6USGH5PMTN5","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"pith_short_16","alias_value":"Y6USGH5PMTN5GVNM","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"pith_short_8","alias_value":"Y6USGH5P","created_at":"2026-07-05T10:15:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Y6USGH5PMTN5GVNMXQAEMHH3BE","target":"record","payload":{"canonical_record":{"source":{"id":"2409.05907","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T15:47:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"1c40c3cde09063ae75c93e0d8b66503f2cc931085f26afc8dcab555ef664294a","abstract_canon_sha256":"e5685333ba0c99c8f10662f44ffd475738d8ad92b7c43a6d6080b31a7ae1f16b"},"schema_version":"1.0"},"canonical_sha256":"c7a9231faf64dbd355acbc00461cfb091d0e65bac6c9a66865d82e52a192d7eb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:15:45.501311Z","signature_b64":"duJHOHMjbtlp4btk9t3VnZT2wMBS+IVDRUtbllJ25AfXr5R+grF+VBSmiPD9rama1rc6D/5xxvK6kJrlht6JDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7a9231faf64dbd355acbc00461cfb091d0e65bac6c9a66865d82e52a192d7eb","last_reissued_at":"2026-07-05T10:15:45.500580Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:15:45.500580Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.05907","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:15:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UHDk6DJF8KVQR4ODbf6zhN8u6TSMamk+09zYdRYLK3f5NUBqoGjKHqCrPwL53JXJ6P7ickEXvsO0WvahaYd3Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T07:38:28.143765Z"},"content_sha256":"5c2160a35a19ae044ce66126effde5f758dea96bd22124496f777fa170769bed","schema_version":"1.0","event_id":"sha256:5c2160a35a19ae044ce66126effde5f758dea96bd22124496f777fa170769bed"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Y6USGH5PMTN5GVNMXQAEMHH3BE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Programming Refusal with Conditional Activation Steering","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Amit Dhurandhar, Bruce W. Lee, Erik Miehling, Inkit Padhi, Karthikeyan Natesan Ramamurthy, Manish Nagireddy, Pierre Dognin","submitted_at":"2024-09-06T15:47:40Z","abstract_excerpt":"LLMs have shown remarkable capabilities, but precisely controlling their response behavior remains challenging. Existing activation steering methods alter LLM behavior indiscriminately, limiting their practical applicability in settings where selective responses are essential, such as content moderation or domain-specific assistants. In this paper, we propose Conditional Activation Steering (CAST), which analyzes LLM activation patterns during inference to selectively apply or withhold activation steering based on the input context. Our method is based on the observation that different categor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.05907","kind":"arxiv","version":3},"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/2409.05907/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":"2026-07-05T10:15:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"P4tsMXTgjuSU4faW9jP8qqWrc8olHcqzTYDjihNFhRP9rv5+RN8lGAbKqoTZyXBrXSt165BpDzGtT45zQNRrDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T07:38:28.144549Z"},"content_sha256":"ea2ae09710ffae91f8861178e1b9f1828024f0875ff986ece1daca311ada440c","schema_version":"1.0","event_id":"sha256:ea2ae09710ffae91f8861178e1b9f1828024f0875ff986ece1daca311ada440c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE/bundle.json","state_url":"https://pith.science/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE/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":"2026-08-17T07:38:28Z","links":{"resolver":"https://pith.science/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE","bundle":"https://pith.science/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE/bundle.json","state":"https://pith.science/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y6USGH5PMTN5GVNMXQAEMHH3BE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Y6USGH5PMTN5GVNMXQAEMHH3BE","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":"e5685333ba0c99c8f10662f44ffd475738d8ad92b7c43a6d6080b31a7ae1f16b","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T15:47:40Z","title_canon_sha256":"1c40c3cde09063ae75c93e0d8b66503f2cc931085f26afc8dcab555ef664294a"},"schema_version":"1.0","source":{"id":"2409.05907","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.05907","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"arxiv_version","alias_value":"2409.05907v3","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.05907","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"pith_short_12","alias_value":"Y6USGH5PMTN5","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"pith_short_16","alias_value":"Y6USGH5PMTN5GVNM","created_at":"2026-07-05T10:15:45Z"},{"alias_kind":"pith_short_8","alias_value":"Y6USGH5P","created_at":"2026-07-05T10:15:45Z"}],"graph_snapshots":[{"event_id":"sha256:ea2ae09710ffae91f8861178e1b9f1828024f0875ff986ece1daca311ada440c","target":"graph","created_at":"2026-07-05T10:15:45Z","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/2409.05907/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLMs have shown remarkable capabilities, but precisely controlling their response behavior remains challenging. Existing activation steering methods alter LLM behavior indiscriminately, limiting their practical applicability in settings where selective responses are essential, such as content moderation or domain-specific assistants. In this paper, we propose Conditional Activation Steering (CAST), which analyzes LLM activation patterns during inference to selectively apply or withhold activation steering based on the input context. Our method is based on the observation that different categor","authors_text":"Amit Dhurandhar, Bruce W. Lee, Erik Miehling, Inkit Padhi, Karthikeyan Natesan Ramamurthy, Manish Nagireddy, Pierre Dognin","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T15:47:40Z","title":"Programming Refusal with Conditional Activation Steering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.05907","kind":"arxiv","version":3},"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":"sha256:5c2160a35a19ae044ce66126effde5f758dea96bd22124496f777fa170769bed","target":"record","created_at":"2026-07-05T10:15:45Z","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":"e5685333ba0c99c8f10662f44ffd475738d8ad92b7c43a6d6080b31a7ae1f16b","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T15:47:40Z","title_canon_sha256":"1c40c3cde09063ae75c93e0d8b66503f2cc931085f26afc8dcab555ef664294a"},"schema_version":"1.0","source":{"id":"2409.05907","kind":"arxiv","version":3}},"canonical_sha256":"c7a9231faf64dbd355acbc00461cfb091d0e65bac6c9a66865d82e52a192d7eb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c7a9231faf64dbd355acbc00461cfb091d0e65bac6c9a66865d82e52a192d7eb","first_computed_at":"2026-07-05T10:15:45.500580Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:15:45.500580Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"duJHOHMjbtlp4btk9t3VnZT2wMBS+IVDRUtbllJ25AfXr5R+grF+VBSmiPD9rama1rc6D/5xxvK6kJrlht6JDg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:15:45.501311Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.05907","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5c2160a35a19ae044ce66126effde5f758dea96bd22124496f777fa170769bed","sha256:ea2ae09710ffae91f8861178e1b9f1828024f0875ff986ece1daca311ada440c"],"state_sha256":"b2b26c516cc79991febefef6f779c0025451a3c00252bd37cb05a18fdd1ed606"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1LbybQM0aHhoYAkXL2auYlC9dmW4fQCaL4CJ8ZhgFZLd3M3Zno6n1z4Wof1BVtQnQbKc9ZP12AG4wDigHfbjCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T07:38:28.153475Z","bundle_sha256":"c273647c9551a38eec38dfda9432a0b77e78ebbd89e0680113bccd36f04a237e"}}