{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.09826", "kind": "paper", "url": "https://arxiv.org/abs/2606.09826"}, {"id": "gh:mxlin043/omnigamearena", "kind": "repo", "url": "https://github.com/mxlin043/OmniGameArena"}], "authors": ["Mingxian Lin", "Shengju Qian", "Yuqi Liu", "Yi-Hua Huang", "Yiyu Wang", "Wei Huang", "Yitang Li", "Fan Zhang", "Zeyu Hu", "Lingting Zhu", "Xin Wang", "Xiaojuan Qi"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.09826", "source_url": "https://arxiv.org/abs/2606.09826"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "OmniGameArena evaluates VLM game agents across twelve Unreal Engine 5 games spanning Solo, PvP, and Coop play. It provides unified action interfaces and two evaluation clocks (PDQ for decision quality, LCRT for real-time latency). The Improvement Dynamics Curve (IDC) measures agent improvement through reflection rounds."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.09826", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.09826", "title": "OmniGameArena"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.09826", "modality": null, "name": "OmniGameArena", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "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_omnigamearena_6afe515a", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "02f41c137e4bffd32c6b3d7561be2bf2db36bab6a59207b9f9ce6e1ee939e53b", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.09826"}, "publisher": null, "released": "2026-06-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.09826", "source_url": "https://arxiv.org/abs/2606.09826"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-09826", "source": "claire_radar", "source_benchmark_id": "2606.09826", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/mxlin043/OmniGameArena", "githubScope": "benchmark_repo", "githubStars": 45, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 19, "hfPaperUrl": "https://huggingface.co/papers/2606.09826", "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": "2026-09-25", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "a05d9d2d817c0f5f2799299c7ebb66c137ba23d2cefb0bb95254e2e1a85a137e", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "OmniGameArena evaluates VLM game agents across twelve Unreal Engine 5 games spanning Solo, PvP, and Coop play. It provides unified action interfaces and two evaluation clocks (PDQ for decision quality, LCRT for real-time latency). The Improvement Dynamics Curve (IDC) measures agent improvement through reflection rounds.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "evaluation protocol evidence"], "snippet": "We address these gaps with OmniGameArena, a real-time benchmark of twelve newly built Unreal Engine 5 games spanning Solo (7), PvP (3), and Coop (2) with unified action interfaces, and the Improvement Dynamics Curve (IDC), an agentic-reflection harness in which a tool-using reflector LLM autonomously refines a bounded skill prompt across multiple rounds."}, "familyId": "bmf_bdedbd5c5172", "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/mxlin043/OmniGameArena", "url": "https://github.com/mxlin043/OmniGameArena"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/mxlin043/OmniGameArena", "url": "https://huggingface.co/papers/2606.09826"}], "scope": "benchmark_repo", "url": "https://github.com/mxlin043/OmniGameArena", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_omnigamearena_6afe515a", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/mxlin043/OmniGameArena", "data": null, "hfPaper": "https://huggingface.co/papers/2606.09826", "paper": "https://arxiv.org/abs/2606.09826", "pdf": "https://arxiv.org/pdf/2606.09826", "project": null, "report": "https://arxiv.org/abs/2606.09826"}, "metrics": [], "motivation": "Vision-language model (VLM) agents are increasingly deployed in interactive game environments.", "name": "OmniGameArena", "oneLine": "OmniGameArena evaluates VLM game agents across twelve Unreal Engine 5 games spanning Solo, PvP, and Coop play. It provides unified action interfaces and two evaluation clocks (PDQ for decision quality, LCRT for real-time latency). The Improvement Dynamics Curve (IDC) measures agent improvement through reflection rounds.", "paperTitle": "OmniGameArena: A Unified UE5 Benchmark for VLM Game Agents with Improvement Dynamics", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.09826", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "OmniGameArena Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/mxlin043/OmniGameArena"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-06-08", "source": {"authors": ["Mingxian Lin", "Shengju Qian", "Yuqi Liu", "Yi-Hua Huang", "Yiyu Wang", "Wei Huang", "Yitang Li", "Fan Zhang", "Zeyu Hu", "Lingting Zhu", "Xin Wang", "Xiaojuan Qi"], "categories": ["cs.CV", "cs.AI"], "id": "2606.09826", "title": "OmniGameArena: A Unified UE5 Benchmark for VLM Game Agents with Improvement Dynamics", "type": "arxiv", "url": "https://arxiv.org/abs/2606.09826"}, "sourceUpdatedAt": "2026-06-08T17:59:43Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Existing game benchmarks report single scores and lack unified protocols for heterogeneous agents. OmniGameArena offers a real-time benchmark with multiple observables, including improvement dynamics, facilitating comparative evaluation of different agent classes."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
