{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.13385", "kind": "paper", "url": "https://arxiv.org/abs/2606.13385"}], "authors": ["Zihao Wang", "Yiming Li", "Yutong Wu", "Kangjie Chen", "Zheyu Liu", "Fok Kar Wai", "Pin-Yu Chen", "Vrizlynn L. L. Thing", "Bo Li", "Dacheng Tao", "Tianwei Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.13385", "source_url": "https://arxiv.org/abs/2606.13385"}, "categories": ["Agents", "Cybersecurity", "Finance", "Language & Knowledge"], "description": {"en": "StakeBench is a stakeholder-centric benchmark for prompt-injection attacks in web agents for online shopping. It decomposes risk into 12 attack objectives across three stakeholder classes, with 264 adversarial cases across 12 product categories."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.13385", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.13385", "title": "StakeBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Data Storage Institute", "Nanyang Technological University", "ST Engineering (Singapore)", "IBM (United States)", "University of Illinois Urbana-Champaign"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.13385", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.13385", "source_url": "https://openalex.org/W7164525138"}, "key": "claire-radar:2606.13385", "modality": null, "name": "StakeBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_stakebench_be4543bc", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "f707db044a21ae9bbfa27a71fd0bfe20eb020b25febae13320b27ed264772394", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.13385"}, "publisher": null, "released": "2026-06-11", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.13385", "source_url": "https://arxiv.org/abs/2606.13385"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-13385", "source": "claire_radar", "source_benchmark_id": "2606.13385", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Cybersecurity", "Finance"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.13385", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "ce4c91daad8cf70bdc63538671da0ef1118b36aa9229478867c967e28a5752e2", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "StakeBench is a stakeholder-centric benchmark for prompt-injection attacks in web agents for online shopping. It decomposes risk into 12 attack objectives across three stakeholder classes, with 264 adversarial cases across 12 product categories.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To capture these properties, we introduce StakeBench, a stakeholder-centric benchmark that systematically categorizes and attributes harm in real-world web agent systems for online shopping."}, "familyId": "bmf_bb61545ac7a2", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_stakebench_be4543bc", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Cybersecurity", "Financial Services"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.13385", "paper": "https://arxiv.org/abs/2606.13385", "pdf": "https://arxiv.org/pdf/2606.13385", "project": null, "report": "https://arxiv.org/abs/2606.13385"}, "metrics": [], "motivation": "LLM-based web agents are increasingly deployed in real-world settings such as e-commerce, where they interact extensively with untrusted web content while executing actions that carry direct financial consequences.", "name": "StakeBench", "oneLine": "StakeBench is a stakeholder-centric benchmark for prompt-injection attacks in web agents for online shopping. It decomposes risk into 12 attack objectives across three stakeholder classes, with 264 adversarial cases across 12 product categories.", "paperTitle": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents", "primaryDomain": "Cybersecurity", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.13385", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-11", "source": {"authors": ["Zihao Wang", "Yiming Li", "Yutong Wu", "Kangjie Chen", "Zheyu Liu", "Fok Kar Wai", "Pin-Yu Chen", "Vrizlynn L. L. Thing", "Bo Li", "Dacheng Tao", "Tianwei Zhang"], "categories": ["cs.CR", "cs.AI", "cs.CY", "cs.HC", "cs.MM"], "id": "2606.13385", "title": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2606.13385"}, "sourceUpdatedAt": "2026-07-27T14:19:24Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Prompt-injection risk is victim-dependent. StakeBench captures asymmetric consequences for different stakeholders, which is overlooked by attack-centric evaluations."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
