{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.03103", "kind": "paper", "url": "https://arxiv.org/abs/2606.03103"}, {"id": "gh:mrwwk/deskcraft", "kind": "repo", "url": "https://github.com/mrwwk/DeskCraft"}], "authors": ["Wenkai Wang", "Tao Xiong", "Jingchen Ni", "Yunpeng Bao", "Xiyun Li", "Tianqi Liu", "Hongcan Guo", "Zilong Huang", "Shengyu Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.03103", "source_url": "https://arxiv.org/abs/2606.03103"}, "categories": ["Agents", "General AI", "Language & Knowledge"], "description": {"en": "DeskCraft evaluates desktop GUI agents on long-horizon professional workflows and human-in-the-loop collaboration across 538 executable tasks in live Ubuntu desktop environments. Covers design, video, audio, and 3D creation software, with execution-based verification."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.03103", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.03103", "title": "DeskCraft"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.03103", "modality": null, "name": "DeskCraft", "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_deskcraft_5fe792c2", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "8f9850c017f35a191639909c88a94cf2a82d5ed8e5a8cbbddeba048dc6368661", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.03103"}, "publisher": null, "released": "2026-06-02", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.03103", "source_url": "https://arxiv.org/abs/2606.03103"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-03103", "source": "claire_radar", "source_benchmark_id": "2606.03103", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/mrwwk/DeskCraft", "githubScope": "benchmark_repo", "githubStars": 94, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 3, "hfPaperUrl": "https://huggingface.co/papers/2606.03103", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "fresh"}, "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": "9197fd4bd53ae02283a924b4d2474a32ca619a045171b712ab56fadefa7887db", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "DeskCraft evaluates desktop GUI agents on long-horizon professional workflows and human-in-the-loop collaboration across 538 executable tasks in live Ubuntu desktop environments. Covers design, video, audio, and 3D creation software, with execution-based verification.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To address this issue, we introduce DeskCraft, a desktop GUI benchmark targeting long horizon creative and engineering workflows and proactive human-agent collaboration."}, "familyId": "bmf_eaf217e35324", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/mrwwk/DeskCraft", "url": "https://github.com/mrwwk/DeskCraft"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/mrwwk/DeskCraft", "url": "https://huggingface.co/papers/2606.03103"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/mrwwk/DeskCraft.", "url": "https://arxiv.org/abs/2606.03103"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/mrwwk/DeskCraft", "url": "https://arxiv.org/abs/2606.03103"}], "scope": "benchmark_repo", "url": "https://github.com/mrwwk/DeskCraft", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_deskcraft_5fe792c2", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/mrwwk/DeskCraft", "data": null, "hfPaper": "https://huggingface.co/papers/2606.03103", "paper": "https://arxiv.org/abs/2606.03103", "pdf": "https://arxiv.org/pdf/2606.03103", "project": null, "report": "https://arxiv.org/abs/2606.03103"}, "metrics": [], "motivation": "Real-world professional desktop workflows in specialized creative and engineering software unfold over long horizons and often require human-in-the-loop coordination, where agents proactively seek necessary information and users provide additional instructions, clarifications, feedback, or corrections as the task progresses.", "name": "DeskCraft", "oneLine": "DeskCraft evaluates desktop GUI agents on long-horizon professional workflows and human-in-the-loop collaboration across 538 executable tasks in live Ubuntu desktop environments. Covers design, video, audio, and 3D creation software, with execution-based verification.", "paperTitle": "DeskCraft: Benchmarking Desktop Agents on Professional Workflows and Human-in-the-Loop Collaboration", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.03103", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "DeskCraft team", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/mrwwk/DeskCraft"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-02", "source": {"authors": ["Wenkai Wang", "Tao Xiong", "Jingchen Ni", "Yunpeng Bao", "Xiyun Li", "Tianqi Liu", "Hongcan Guo", "Zilong Huang", "Shengyu Zhang"], "categories": ["cs.AI"], "id": "2606.03103", "title": "DeskCraft: Benchmarking Desktop Agents on Professional Workflows and Human-in-the-Loop Collaboration", "type": "arxiv", "url": "https://arxiv.org/abs/2606.03103"}, "sourceUpdatedAt": "2026-06-02T03:42:34Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Existing desktop benchmarks simplify tasks and lack human-agent interaction. DeskCraft measures agent performance on realistic workflows and proactive collaboration, identifying gaps in long-horizon delivery and clarification."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
