{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SAOKXUIROZLSUSNT266VSUX4RP","short_pith_number":"pith:SAOKXUIR","schema_version":"1.0","canonical_sha256":"901cabd11176572a49b3d7bd5952fc8bd0eeb6916de54b0f965b0c23bbfd99f2","source":{"kind":"arxiv","id":"2605.23220","version":1},"attestation_state":"computed","paper":{"title":"WMAttack: Automated Attack Search for Adversarial Evaluation of World-Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andras Balogh, Cheng Guo, Dacheng Tao, Mark Jelasity, Shi Fu, Siyuan Liang, Zhixiang Guo","submitted_at":"2026-05-22T04:31:09Z","abstract_excerpt":"Despite the growing use of world models as decision-making agents, their adversarial robustness remains underexplored due to the lack of dedicated automated evaluation methods. A key obstacle is that attack evaluation must be both accurate and efficient: weak manually tuned attacks can overestimate robustness, while exhaustive hyperparameter search is prohibitively expensive because each candidate requires closed-loop rollouts through learned latent dynamics. We introduce WMAttack, an automated attack-search framework for adversarial evaluation of world-model agents. WMAttack formulates robust"},"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":"2605.23220","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-22T04:31:09Z","cross_cats_sorted":[],"title_canon_sha256":"24451d952d481e06e455f2186b0372977a011f701f1c575734cbdee4c5bf4c52","abstract_canon_sha256":"0b6f522355437f6ad5c2be202d853dacc296bd7d1b7d84727f8bbbfe26015687"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:44.361346Z","signature_b64":"WUpW7jZQAZvul61iqtoAay+wPuyopyjKJr6KAnZsZLg7AhaSx7lW/V5G16EVpE6QqHzRpbI24VIdcFKFzFYYBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"901cabd11176572a49b3d7bd5952fc8bd0eeb6916de54b0f965b0c23bbfd99f2","last_reissued_at":"2026-05-25T02:01:44.360613Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:44.360613Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WMAttack: Automated Attack Search for Adversarial Evaluation of World-Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andras Balogh, Cheng Guo, Dacheng Tao, Mark Jelasity, Shi Fu, Siyuan Liang, Zhixiang Guo","submitted_at":"2026-05-22T04:31:09Z","abstract_excerpt":"Despite the growing use of world models as decision-making agents, their adversarial robustness remains underexplored due to the lack of dedicated automated evaluation methods. A key obstacle is that attack evaluation must be both accurate and efficient: weak manually tuned attacks can overestimate robustness, while exhaustive hyperparameter search is prohibitively expensive because each candidate requires closed-loop rollouts through learned latent dynamics. We introduce WMAttack, an automated attack-search framework for adversarial evaluation of world-model agents. WMAttack formulates robust"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23220","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/2605.23220/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":"2605.23220","created_at":"2026-05-25T02:01:44.360741+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.23220v1","created_at":"2026-05-25T02:01:44.360741+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23220","created_at":"2026-05-25T02:01:44.360741+00:00"},{"alias_kind":"pith_short_12","alias_value":"SAOKXUIROZLS","created_at":"2026-05-25T02:01:44.360741+00:00"},{"alias_kind":"pith_short_16","alias_value":"SAOKXUIROZLSUSNT","created_at":"2026-05-25T02:01:44.360741+00:00"},{"alias_kind":"pith_short_8","alias_value":"SAOKXUIR","created_at":"2026-05-25T02:01:44.360741+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/SAOKXUIROZLSUSNT266VSUX4RP","json":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP.json","graph_json":"https://pith.science/api/pith-number/SAOKXUIROZLSUSNT266VSUX4RP/graph.json","events_json":"https://pith.science/api/pith-number/SAOKXUIROZLSUSNT266VSUX4RP/events.json","paper":"https://pith.science/paper/SAOKXUIR"},"agent_actions":{"view_html":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP","download_json":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP.json","view_paper":"https://pith.science/paper/SAOKXUIR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.23220&json=true","fetch_graph":"https://pith.science/api/pith-number/SAOKXUIROZLSUSNT266VSUX4RP/graph.json","fetch_events":"https://pith.science/api/pith-number/SAOKXUIROZLSUSNT266VSUX4RP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP/action/storage_attestation","attest_author":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP/action/author_attestation","sign_citation":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP/action/citation_signature","submit_replication":"https://pith.science/pith/SAOKXUIROZLSUSNT266VSUX4RP/action/replication_record"}},"created_at":"2026-05-25T02:01:44.360741+00:00","updated_at":"2026-05-25T02:01:44.360741+00:00"}