{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XORO64FWH3NZZLJUYUR7ACHTOR","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":"0103d671f39a7e383b2d785bf2bcb0820f908dc44677cd364ad324c8435de45f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-14T11:15:27Z","title_canon_sha256":"50ffa1e2dbc504b2f3a8a98321950c38bb2dc2f576614730e654dcc423ef1663"},"schema_version":"1.0","source":{"id":"2502.10487","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.10487","created_at":"2026-07-05T10:14:53Z"},{"alias_kind":"arxiv_version","alias_value":"2502.10487v1","created_at":"2026-07-05T10:14:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.10487","created_at":"2026-07-05T10:14:53Z"},{"alias_kind":"pith_short_12","alias_value":"XORO64FWH3NZ","created_at":"2026-07-05T10:14:53Z"},{"alias_kind":"pith_short_16","alias_value":"XORO64FWH3NZZLJU","created_at":"2026-07-05T10:14:53Z"},{"alias_kind":"pith_short_8","alias_value":"XORO64FW","created_at":"2026-07-05T10:14:53Z"}],"graph_snapshots":[{"event_id":"sha256:467e47bafe84d80bef7c216fc74d002b8c2615c178ac0765165eced6dac42275","target":"graph","created_at":"2026-07-05T10:14:53Z","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/2502.10487/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating the robustness of LLMs to adversarial attacks is crucial for safe deployment, yet current red-teaming methods are often prohibitively expensive. We compare the ability of fast proxy metrics to predict the real-world robustness of an LLM against a simulated attacker ensemble. This allows us to estimate a model's robustness to computationally expensive attacks without requiring runs of the attacks themselves. Specifically, we consider gradient-descent-based embedding-space attacks, prefilling attacks, and direct prompting. Even though direct prompting in particular does not achieve hi","authors_text":"Jan Schuchardt, Leo Schwinn, Stephan G\\\"unnemann, Tim Beyer","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-14T11:15:27Z","title":"Fast Proxies for LLM Robustness Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.10487","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":"sha256:2575f675889d279c6deda3833ab622d8db26e90536f412e1ab525627b49c0b29","target":"record","created_at":"2026-07-05T10:14:53Z","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":"0103d671f39a7e383b2d785bf2bcb0820f908dc44677cd364ad324c8435de45f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-14T11:15:27Z","title_canon_sha256":"50ffa1e2dbc504b2f3a8a98321950c38bb2dc2f576614730e654dcc423ef1663"},"schema_version":"1.0","source":{"id":"2502.10487","kind":"arxiv","version":1}},"canonical_sha256":"bba2ef70b63edb9cad34c523f008f3746c1953ea127aa8e7e41b5bd2f070c1c1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bba2ef70b63edb9cad34c523f008f3746c1953ea127aa8e7e41b5bd2f070c1c1","first_computed_at":"2026-07-05T10:14:53.481625Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:14:53.481625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"C84VEHBiC6R6AuygqhM/SFr4irxgzlkPOfzcGK4fqWf+71s3kUnxV2OLGTuSaegOatKBrLnfnmu+YAS7dhh0DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:14:53.482124Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.10487","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2575f675889d279c6deda3833ab622d8db26e90536f412e1ab525627b49c0b29","sha256:467e47bafe84d80bef7c216fc74d002b8c2615c178ac0765165eced6dac42275"],"state_sha256":"5e2cf10cdc14a58a97c919960c7d6b8173f6173a753beed5b77a349aeb93b9c4"}