{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:R6UZVWFAZ5NHQX2TIVOZC3YQ5B","short_pith_number":"pith:R6UZVWFA","canonical_record":{"source":{"id":"2512.06655","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-07T04:46:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"90a57ab8b5a71a55f6cf56e4f457cbf66a46496e1556c5b8bd6ce43498f5a7d8","abstract_canon_sha256":"f656efedb389559f73c5f3f2a5e02b7824cad429330a3995c2618430c32cf2ef"},"schema_version":"1.0"},"canonical_sha256":"8fa99ad8a0cf5a785f53455d916f10e8549258efe51e4c45ce3834eee02e8703","source":{"kind":"arxiv","id":"2512.06655","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.06655","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"arxiv_version","alias_value":"2512.06655v3","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.06655","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"pith_short_12","alias_value":"R6UZVWFAZ5NH","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"pith_short_16","alias_value":"R6UZVWFAZ5NHQX2T","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"pith_short_8","alias_value":"R6UZVWFA","created_at":"2026-05-20T00:00:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:R6UZVWFAZ5NHQX2TIVOZC3YQ5B","target":"record","payload":{"canonical_record":{"source":{"id":"2512.06655","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-07T04:46:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"90a57ab8b5a71a55f6cf56e4f457cbf66a46496e1556c5b8bd6ce43498f5a7d8","abstract_canon_sha256":"f656efedb389559f73c5f3f2a5e02b7824cad429330a3995c2618430c32cf2ef"},"schema_version":"1.0"},"canonical_sha256":"8fa99ad8a0cf5a785f53455d916f10e8549258efe51e4c45ce3834eee02e8703","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:29.554240Z","signature_b64":"EAaVVk1uS41qk4wEeYv42xQBg/d7dkeM4m2zm+8N1leWb6QKyQjox1Azkzp+BIRj7R202y/AeiQv2CYxvsflDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8fa99ad8a0cf5a785f53455d916f10e8549258efe51e4c45ce3834eee02e8703","last_reissued_at":"2026-05-20T00:00:29.553709Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:29.553709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2512.06655","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":"2026-05-20T00:00:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p/vQ8ivsav6PzomV2+H5UCL34fv1jaw/DSowm366voBzgiOVzcJdm9+oLx96I58YsktLTFkjv1UryuXAdo4DBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T01:22:57.329100Z"},"content_sha256":"18e36acacd22e9b38c54b3d5a88839d60352ffee435545833e09fbaedcacc0c0","schema_version":"1.0","event_id":"sha256:18e36acacd22e9b38c54b3d5a88839d60352ffee435545833e09fbaedcacc0c0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:R6UZVWFAZ5NHQX2TIVOZC3YQ5B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Graph-Regularized Sparse Autoencoders for LLM Safety Steering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Federico Cinus, Jehyeok Yeon, Luca Luceri, Yifan Wu","submitted_at":"2025-12-07T04:46:30Z","abstract_excerpt":"Sparse autoencoders (SAEs) are increasingly used to extract activation directions for inference-time steering, but their standard sparsity objective treats latent features as independent. This prior can be poorly matched to high-level safety behaviors, where refusal and harmful compliance appear to depend on distributed structure in activation space. We introduce Graph-Regularized Sparse Autoencoders (GSAE), a dictionary-learning method that learns safety-steering directions by smoothing SAE decoder vectors over a neuron co-activation graph and applying the resulting direction bank through a t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.06655","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/2512.06655/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":"2026-05-20T00:00:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kfvY6ElmumQ6IiYsuvUxw+t/asd3wBd5Ibsgxbx+AWpj1dBh/Y00ic7Lq6e+M9x/SCTzHIJWm5Zp3RNFZ24lBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T01:22:57.329503Z"},"content_sha256":"b6454fc053cc2071c8d35631ac4220ce042bfa63471f286e0490c0df6829ae66","schema_version":"1.0","event_id":"sha256:b6454fc053cc2071c8d35631ac4220ce042bfa63471f286e0490c0df6829ae66"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B/bundle.json","state_url":"https://pith.science/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B/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":"2026-05-27T01:22:57Z","links":{"resolver":"https://pith.science/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B","bundle":"https://pith.science/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B/bundle.json","state":"https://pith.science/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R6UZVWFAZ5NHQX2TIVOZC3YQ5B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:R6UZVWFAZ5NHQX2TIVOZC3YQ5B","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":"f656efedb389559f73c5f3f2a5e02b7824cad429330a3995c2618430c32cf2ef","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-07T04:46:30Z","title_canon_sha256":"90a57ab8b5a71a55f6cf56e4f457cbf66a46496e1556c5b8bd6ce43498f5a7d8"},"schema_version":"1.0","source":{"id":"2512.06655","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.06655","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"arxiv_version","alias_value":"2512.06655v3","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.06655","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"pith_short_12","alias_value":"R6UZVWFAZ5NH","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"pith_short_16","alias_value":"R6UZVWFAZ5NHQX2T","created_at":"2026-05-20T00:00:29Z"},{"alias_kind":"pith_short_8","alias_value":"R6UZVWFA","created_at":"2026-05-20T00:00:29Z"}],"graph_snapshots":[{"event_id":"sha256:b6454fc053cc2071c8d35631ac4220ce042bfa63471f286e0490c0df6829ae66","target":"graph","created_at":"2026-05-20T00:00:29Z","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/2512.06655/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Sparse autoencoders (SAEs) are increasingly used to extract activation directions for inference-time steering, but their standard sparsity objective treats latent features as independent. This prior can be poorly matched to high-level safety behaviors, where refusal and harmful compliance appear to depend on distributed structure in activation space. We introduce Graph-Regularized Sparse Autoencoders (GSAE), a dictionary-learning method that learns safety-steering directions by smoothing SAE decoder vectors over a neuron co-activation graph and applying the resulting direction bank through a t","authors_text":"Federico Cinus, Jehyeok Yeon, Luca Luceri, Yifan Wu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-07T04:46:30Z","title":"Graph-Regularized Sparse Autoencoders for LLM Safety Steering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.06655","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":"sha256:18e36acacd22e9b38c54b3d5a88839d60352ffee435545833e09fbaedcacc0c0","target":"record","created_at":"2026-05-20T00:00:29Z","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":"f656efedb389559f73c5f3f2a5e02b7824cad429330a3995c2618430c32cf2ef","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-07T04:46:30Z","title_canon_sha256":"90a57ab8b5a71a55f6cf56e4f457cbf66a46496e1556c5b8bd6ce43498f5a7d8"},"schema_version":"1.0","source":{"id":"2512.06655","kind":"arxiv","version":3}},"canonical_sha256":"8fa99ad8a0cf5a785f53455d916f10e8549258efe51e4c45ce3834eee02e8703","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8fa99ad8a0cf5a785f53455d916f10e8549258efe51e4c45ce3834eee02e8703","first_computed_at":"2026-05-20T00:00:29.553709Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:00:29.553709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EAaVVk1uS41qk4wEeYv42xQBg/d7dkeM4m2zm+8N1leWb6QKyQjox1Azkzp+BIRj7R202y/AeiQv2CYxvsflDQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:00:29.554240Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.06655","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:18e36acacd22e9b38c54b3d5a88839d60352ffee435545833e09fbaedcacc0c0","sha256:b6454fc053cc2071c8d35631ac4220ce042bfa63471f286e0490c0df6829ae66"],"state_sha256":"120447b0f4d3cd4e3f118073b52a2728dd33204942d8644d4d4d696e9a83ac32"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5VQqk5NENf+IB1ZVVcnpKgWD5e6iV+4UUSqOxSn+hMnn75KKaV928+Eo5uMnKWqTSlaiu71SU7ZVVosCPFi4Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T01:22:57.332371Z","bundle_sha256":"4883a0766cc31abe8bccef5fb9a9cb12b0dc707cdfda10c14c471179fddec1cf"}}