{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NNYZERLHDMG4I4RNNK2SP4GHDD","short_pith_number":"pith:NNYZERLH","schema_version":"1.0","canonical_sha256":"6b719245671b0dc4722d6ab527f0c718c8e932a1a3d16772755789333f99010a","source":{"kind":"arxiv","id":"2605.25707","version":1},"attestation_state":"computed","paper":{"title":"AgentHijack: Benchmarking Computer Use Agent Robustness to Common Environment Corruptions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Han, Jianing Zhu, Jingwei Sun, Tongliang Liu, Xia Hu, Yuanyi Li","submitted_at":"2026-05-25T11:09:22Z","abstract_excerpt":"Autonomous computer use agents that powered by multimodal large language models (MLLMs) are emerging as capable assistants for completing complex digital workflows. However, real-world execution environments are far from ideal: pop-ups, resolution changes, and competing applications frequently interfere with agent perception and control. We introduce AgentHijack, a benchmark designed to evaluate the robustness of computer-use agents under common corruptions, where the uncertainties in dynamic environment disrupt the execution flow without direct adversarial intent. Specifically, AgentHijack in"},"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":"2605.25707","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-25T11:09:22Z","cross_cats_sorted":[],"title_canon_sha256":"f4613692d8509c784a28f7a5e7f3f2439efa332ed20b36ab57acb8dbf46ce47d","abstract_canon_sha256":"731ebad34339c11a90148a5a81b795122d8277291cdb6c8c59e9be0950a3c4e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:51.510836Z","signature_b64":"ZUWcSYe5ntZCSYUZIeD06giMrxwuD5c/vizhRtt1MAiUV1N5K3Cvi+O5X+fSMm2WMXLpvUtm/FMU/48rJXbSBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b719245671b0dc4722d6ab527f0c718c8e932a1a3d16772755789333f99010a","last_reissued_at":"2026-05-26T02:04:51.510279Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:51.510279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AgentHijack: Benchmarking Computer Use Agent Robustness to Common Environment Corruptions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Han, Jianing Zhu, Jingwei Sun, Tongliang Liu, Xia Hu, Yuanyi Li","submitted_at":"2026-05-25T11:09:22Z","abstract_excerpt":"Autonomous computer use agents that powered by multimodal large language models (MLLMs) are emerging as capable assistants for completing complex digital workflows. However, real-world execution environments are far from ideal: pop-ups, resolution changes, and competing applications frequently interfere with agent perception and control. We introduce AgentHijack, a benchmark designed to evaluate the robustness of computer-use agents under common corruptions, where the uncertainties in dynamic environment disrupt the execution flow without direct adversarial intent. Specifically, AgentHijack in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.25707","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/2605.25707/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":"2605.25707","created_at":"2026-05-26T02:04:51.510366+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.25707v1","created_at":"2026-05-26T02:04:51.510366+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.25707","created_at":"2026-05-26T02:04:51.510366+00:00"},{"alias_kind":"pith_short_12","alias_value":"NNYZERLHDMG4","created_at":"2026-05-26T02:04:51.510366+00:00"},{"alias_kind":"pith_short_16","alias_value":"NNYZERLHDMG4I4RN","created_at":"2026-05-26T02:04:51.510366+00:00"},{"alias_kind":"pith_short_8","alias_value":"NNYZERLH","created_at":"2026-05-26T02:04:51.510366+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/NNYZERLHDMG4I4RNNK2SP4GHDD","json":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD.json","graph_json":"https://pith.science/api/pith-number/NNYZERLHDMG4I4RNNK2SP4GHDD/graph.json","events_json":"https://pith.science/api/pith-number/NNYZERLHDMG4I4RNNK2SP4GHDD/events.json","paper":"https://pith.science/paper/NNYZERLH"},"agent_actions":{"view_html":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD","download_json":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD.json","view_paper":"https://pith.science/paper/NNYZERLH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.25707&json=true","fetch_graph":"https://pith.science/api/pith-number/NNYZERLHDMG4I4RNNK2SP4GHDD/graph.json","fetch_events":"https://pith.science/api/pith-number/NNYZERLHDMG4I4RNNK2SP4GHDD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD/action/storage_attestation","attest_author":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD/action/author_attestation","sign_citation":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD/action/citation_signature","submit_replication":"https://pith.science/pith/NNYZERLHDMG4I4RNNK2SP4GHDD/action/replication_record"}},"created_at":"2026-05-26T02:04:51.510366+00:00","updated_at":"2026-05-26T02:04:51.510366+00:00"}