{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.24597", "kind": "paper", "url": "https://arxiv.org/abs/2606.24597"}, {"id": "gh:qwenlm/qwen-agentworld", "kind": "repo", "url": "https://github.com/QwenLM/Qwen-AgentWorld"}, {"id": "hf:qwen/agentworldbench", "kind": "dataset", "url": "https://huggingface.co/datasets/Qwen/AgentWorldBench"}], "authors": ["Yuxin Zuo", "Zikai Xiao", "Li Sheng", "Fei Huang", "Jianhong Tu", "Yuxuan Liu", "Tianyi Tang", "Xiaomeng Hu", "Yang Su", "Qingfeng Lan", "Yantao Liu", "Qin Zhu", "Yinger Zhang", "Bowen Yu", "Haiquan Zhao", "Haiyang Xu", "Jianxin Yang", "Jiayang Cheng", "Junyang Wang", "Lianghao Deng", "Mingfeng Xue", "Tianyi Bai", "Yang Fan", "Yubo Ma", "Yucheng Li", "Zeyu Cui", "Zhihai Wang", "Zhihui Xie", "Zhuorui Ye", "An Yang", "Dayiheng Liu", "Jingren Zhou", "Ning Ding"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.24597", "source_url": "https://arxiv.org/abs/2606.24597"}, "categories": ["General AI", "Language & Knowledge", "cs.CL"], "description": {"en": "AgentWorldBench evaluates language world models on simulation fidelity across 7 domains (MCP, Search, Terminal, SWE, Android, Web, OS) using real-world trajectories and rubric-based scoring."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.24597", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.24597", "title": "AgentWorldBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.24597", "modality": null, "name": "AgentWorldBench", "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_agentworldbench_775e3273", "origin_source": "arxiv", "recognition_confidence": "0.95", "record_sha256": "5d2a7d34d8a6696dd430519929b0aeda707bdab699fd4638b91da9b2bd011c43", "relation": "introduces", "review_state": "source-reviewed", "reviewed_at": "2026-08-30", "source_url": "https://arxiv.org/abs/2606.24597"}, "publisher": null, "released": "2026-06-23", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.24597", "source_url": "https://arxiv.org/abs/2606.24597"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-24597", "source": "claire_radar", "source_benchmark_id": "2606.24597", "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/QwenLM/Qwen-AgentWorld", "githubScope": "benchmark_repo", "githubStars": 1025, "hfDailySubmittedAt": null, "hfDataset": "Qwen/AgentWorldBench", "hfDatasetDownloads": 958, "hfDatasetLikes": 109, "hfPaperUpvotes": 167, "hfPaperUrl": "https://huggingface.co/papers/2606.24597", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Interactive Environment", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "d2f0149d859e2cc3718f357e72bdc1a4b58e0d7c89f4f4e6cdffa38eb0cc7176", "model": "deepseek-v4-flash"}, "curation": {"reviewedAt": "2026-08-30", "sources": ["https://arxiv.org/abs/2606.24597", "https://github.com/QwenLM/Qwen-AgentWorld", "https://huggingface.co/datasets/Qwen/AgentWorldBench"], "state": "source-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "AgentWorldBench evaluates language world models on simulation fidelity across 7 domains (MCP, Search, Terminal, SWE, Android, Web, OS) using real-world trajectories and rubric-based scoring.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "To evaluate language world models, we present AgentWorldBench, a comprehensive benchmark constructed from real-world interactions of 5 frontier models on 9 established benchmarks."}, "familyId": "bmf_2508b6fd7ca7", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 1, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/QwenLM/Qwen-AgentWorld", "url": "https://github.com/QwenLM/Qwen-AgentWorld"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/QwenLM/Qwen-AgentWorld", "url": "https://huggingface.co/papers/2606.24597"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/QwenLM/Qwen-AgentWorld", "url": "https://huggingface.co/api/papers/2606.24597"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/QwenLM/Qwen-AgentWorld", "url": "https://arxiv.org/abs/2606.24597"}, {"basis": "linked-from-data", "linkedUrl": "https://github.com/QwenLM/Qwen-AgentWorld/tree/main/eval", "url": "https://huggingface.co/datasets/Qwen/AgentWorldBench"}, {"basis": "linked-from-data", "linkedUrl": "https://github.com/QwenLM/Qwen-AgentWorld", "url": "https://huggingface.co/datasets/Qwen/AgentWorldBench"}], "scope": "project_repo", "url": "https://github.com/QwenLM/Qwen-AgentWorld", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_agentworldbench_775e3273", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/QwenLM/Qwen-AgentWorld", "data": "https://huggingface.co/datasets/Qwen/AgentWorldBench", "hfPaper": "https://huggingface.co/papers/2606.24597", "paper": "https://arxiv.org/abs/2606.24597", "pdf": "https://arxiv.org/pdf/2606.24597", "project": null, "report": "https://arxiv.org/abs/2606.24597"}, "metrics": [], "motivation": "A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning.", "name": "AgentWorldBench", "oneLine": "AgentWorldBench evaluates language world models on simulation fidelity across 7 domains (MCP, Search, Terminal, SWE, Android, Web, OS) using real-world trajectories and rubric-based scoring.", "paperTitle": "Qwen-AgentWorld: Language World Models for General Agents", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.24597", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 0.95, "relation": "introduces", "releasedAt": "2026-06-23", "source": {"authors": ["Yuxin Zuo", "Zikai Xiao", "Li Sheng", "Fei Huang", "Jianhong Tu", "Yuxuan Liu", "Tianyi Tang", "Xiaomeng Hu", "Yang Su", "Qingfeng Lan", "Yantao Liu", "Qin Zhu", "Yinger Zhang", "Bowen Yu", "Haiquan Zhao", "Haiyang Xu", "Jianxin Yang", "Jiayang Cheng", "Junyang Wang", "Lianghao Deng", "Mingfeng Xue", "Tianyi Bai", "Yang Fan", "Yubo Ma", "Yucheng Li", "Zeyu Cui", "Zhihai Wang", "Zhihui Xie", "Zhuorui Ye", "An Yang", "Dayiheng Liu", "Jingren Zhou", "Ning Ding"], "categories": ["cs.CL"], "id": "2606.24597", "title": "Qwen-AgentWorld: Language World Models for General Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2606.24597"}, "sourceUpdatedAt": "2026-06-23T13:53:55Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "This benchmark addresses the lack of systematic evaluation for language world models in agentic environments, providing a standardized protocol that enables direct comparison and guides development of general agents."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
