{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4F23EYHKXPFOMBTKW4HNBKM6AW","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":"8e0a8376a003662e2fb7135d69fc712a66c0af60f0d1f0fac79a25349a0ee2bb","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T15:51:23Z","title_canon_sha256":"43886182bde026e40082aec1b28b2f260df3c50a0b4e81afa82d6de0a2ec27b1"},"schema_version":"1.0","source":{"id":"2507.20333","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.20333","created_at":"2026-07-05T11:44:17Z"},{"alias_kind":"arxiv_version","alias_value":"2507.20333v1","created_at":"2026-07-05T11:44:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.20333","created_at":"2026-07-05T11:44:17Z"},{"alias_kind":"pith_short_12","alias_value":"4F23EYHKXPFO","created_at":"2026-07-05T11:44:17Z"},{"alias_kind":"pith_short_16","alias_value":"4F23EYHKXPFOMBTK","created_at":"2026-07-05T11:44:17Z"},{"alias_kind":"pith_short_8","alias_value":"4F23EYHK","created_at":"2026-07-05T11:44:17Z"}],"graph_snapshots":[{"event_id":"sha256:bb985a5593f519eb76aa386e41ee6bc63748bcec24ca2d1e3f743ab7104089e5","target":"graph","created_at":"2026-07-05T11:44:17Z","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/2507.20333/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The focus on safety alignment in large language models (LLMs) has increased significantly due to their widespread adoption across different domains. The scale of LLMs play a contributing role in their success, and the growth in parameter count follows larger hidden dimensions. In this paper, we hypothesize that while the increase in dimensions has been a key advantage, it may lead to emergent problems as well. These problems emerge as the linear structures in the activation space can be exploited, in the form of activation engineering, to circumvent its safety alignment. Through detailed visua","authors_text":"Laziz U. Abdullaev, Rachel S.Y. Teo, Tan M. Nguyen","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T15:51:23Z","title":"The Blessing and Curse of Dimensionality in Safety Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.20333","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:cc9bf3d3b4f2fbe4af7a785026d3bb854f6e7b098cb4508f71a3799bbd750624","target":"record","created_at":"2026-07-05T11:44:17Z","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":"8e0a8376a003662e2fb7135d69fc712a66c0af60f0d1f0fac79a25349a0ee2bb","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T15:51:23Z","title_canon_sha256":"43886182bde026e40082aec1b28b2f260df3c50a0b4e81afa82d6de0a2ec27b1"},"schema_version":"1.0","source":{"id":"2507.20333","kind":"arxiv","version":1}},"canonical_sha256":"e175b260eabbcae6066ab70ed0a99e05a729988cd83d77f33f92bb2c82558fe9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e175b260eabbcae6066ab70ed0a99e05a729988cd83d77f33f92bb2c82558fe9","first_computed_at":"2026-07-05T11:44:17.143624Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:44:17.143624Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CRVY9aTiQMaXxGiJ3g2lsToYt0K0jPjJWM+YqaCozf4A1J5m7/nsoxX6HwMY5IXnuX81jeEGb4kriGV8T9xCCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:44:17.144082Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.20333","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cc9bf3d3b4f2fbe4af7a785026d3bb854f6e7b098cb4508f71a3799bbd750624","sha256:bb985a5593f519eb76aa386e41ee6bc63748bcec24ca2d1e3f743ab7104089e5"],"state_sha256":"dd3909715d222e62c73746c52a583e0472be8634e4cff392616444b021ec3bfa"}