{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2QSZEU4DQZ3AODKS6TOCIXIYGA","short_pith_number":"pith:2QSZEU4D","schema_version":"1.0","canonical_sha256":"d4259253838676070d52f4dc245d18302e57b909f82e5cbc57afe57e4bf0c47d","source":{"kind":"arxiv","id":"2606.11409","version":1},"attestation_state":"computed","paper":{"title":"Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Bogl\\'arka Ecsedi, Colin Raffel, Malikeh Ehghaghi, Marsha Chechik","submitted_at":"2026-06-09T19:59:12Z","abstract_excerpt":"Adversarial robustness evaluations of large language models (LLMs) typically report attack success rate (ASR) under fixed query budgets, implicitly treating all attacks as equally costly. In practice, the computational expense of different attack strategies can vary by orders of magnitude. Consequently, ASR at a fixed budget can obscure the true effort required to jailbreak a model, thereby making it hard to determine whether an attack's cost justifies its payoff to the attacker. We propose a compute-aware evaluation framework based on computational pressure, measured in cumulative floating-po"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.11409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-09T19:59:12Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"700d553c74f2139d45dd3bd1314849ddd174edc11a5ad0883368b13e6d165d8f","abstract_canon_sha256":"eede5c6257f17e2d4ce397111440015f75b94c41ace21b037b09a9f0142a94f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:47.101937Z","signature_b64":"hXnOs5A4qdX5LqvZaJsuljOSAdhiH26GddQowTqD87GkRnS2j4uTgticAyxzBbzT0g0IQHPLXthDQMmKgIJkDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4259253838676070d52f4dc245d18302e57b909f82e5cbc57afe57e4bf0c47d","last_reissued_at":"2026-06-11T01:09:47.100930Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:47.100930Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Bogl\\'arka Ecsedi, Colin Raffel, Malikeh Ehghaghi, Marsha Chechik","submitted_at":"2026-06-09T19:59:12Z","abstract_excerpt":"Adversarial robustness evaluations of large language models (LLMs) typically report attack success rate (ASR) under fixed query budgets, implicitly treating all attacks as equally costly. In practice, the computational expense of different attack strategies can vary by orders of magnitude. Consequently, ASR at a fixed budget can obscure the true effort required to jailbreak a model, thereby making it hard to determine whether an attack's cost justifies its payoff to the attacker. We propose a compute-aware evaluation framework based on computational pressure, measured in cumulative floating-po"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11409","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/2606.11409/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.11409","created_at":"2026-06-11T01:09:47.101067+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.11409v1","created_at":"2026-06-11T01:09:47.101067+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11409","created_at":"2026-06-11T01:09:47.101067+00:00"},{"alias_kind":"pith_short_12","alias_value":"2QSZEU4DQZ3A","created_at":"2026-06-11T01:09:47.101067+00:00"},{"alias_kind":"pith_short_16","alias_value":"2QSZEU4DQZ3AODKS","created_at":"2026-06-11T01:09:47.101067+00:00"},{"alias_kind":"pith_short_8","alias_value":"2QSZEU4D","created_at":"2026-06-11T01:09:47.101067+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA","json":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA.json","graph_json":"https://pith.science/api/pith-number/2QSZEU4DQZ3AODKS6TOCIXIYGA/graph.json","events_json":"https://pith.science/api/pith-number/2QSZEU4DQZ3AODKS6TOCIXIYGA/events.json","paper":"https://pith.science/paper/2QSZEU4D"},"agent_actions":{"view_html":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA","download_json":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA.json","view_paper":"https://pith.science/paper/2QSZEU4D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.11409&json=true","fetch_graph":"https://pith.science/api/pith-number/2QSZEU4DQZ3AODKS6TOCIXIYGA/graph.json","fetch_events":"https://pith.science/api/pith-number/2QSZEU4DQZ3AODKS6TOCIXIYGA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA/action/storage_attestation","attest_author":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA/action/author_attestation","sign_citation":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA/action/citation_signature","submit_replication":"https://pith.science/pith/2QSZEU4DQZ3AODKS6TOCIXIYGA/action/replication_record"}},"created_at":"2026-06-11T01:09:47.101067+00:00","updated_at":"2026-06-11T01:09:47.101067+00:00"}