{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.11216", "kind": "paper", "url": "https://arxiv.org/abs/2608.11216"}, {"id": "gh:electronicarts/autoworldmodelbench", "kind": "repo", "url": "https://github.com/electronicarts/AutoWorldModelBench"}, {"kind": "website", "url": "https://electronicarts.github.io/AutoWorldModelBench/"}], "authors": ["Marjan Moodi", "Xuankang Zhu", "Fernando De Mesentier Silva", "Harold Chaput", "Mohammad Reza Taesiri"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.11216", "source_url": "https://arxiv.org/abs/2608.11216"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "AutoWorldModel-Bench evaluates coding agents on autonomously improving a base world model under fixed compute budget across eight game environments. Uses structured-state representation, with held-out test split and closed-loop iteration."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.11216", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.11216", "title": "AutoWorldModel-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Electronic Arts (United Kingdom)", "Electronic Arts (United States)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.11216", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.11216", "source_url": "https://openalex.org/W7202372866"}, "key": "claire-radar:2608.11216", "modality": null, "name": "AutoWorldModel-Bench", "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_autoworldmodel-bench_01d33f6e", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "6a21e7c0376d6d980edf811706bcbfc4257cc9352fb8374781abe998ca187a63", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.11216"}, "publisher": null, "released": "2026-07-20", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.11216", "source_url": "https://arxiv.org/abs/2608.11216"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-11216", "source": "claire_radar", "source_benchmark_id": "2608.11216", "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/electronicarts/AutoWorldModelBench", "githubScope": "benchmark_repo", "githubStars": 4, "hfDailySubmittedAt": "2026-08-13T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 14, "hfPaperUrl": "https://huggingface.co/papers/2608.11216", "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": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "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": "5494aae2e8fbd6ee1e24e3e85e999963339b379d2b11d751c975eac82f277ae8", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "AutoWorldModel-Bench evaluates coding agents on autonomously improving a base world model under fixed compute budget across eight game environments. Uses structured-state representation, with held-out test split and closed-loop iteration.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce AutoWorldModel-Bench, a closed-loop benchmark in which frontier coding agents autonomously improve a provided base world model under a fixed compute budget."}, "familyId": "bmf_c9aa949f276e", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "linked-from-project", "linkedUrl": "https://github.com/electronicarts/AutoWorldModelBench", "url": "https://electronicarts.github.io/AutoWorldModelBench/"}], "scope": "benchmark_repo", "url": "https://github.com/electronicarts/AutoWorldModelBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_autoworldmodel-bench_01d33f6e", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/electronicarts/AutoWorldModelBench", "data": null, "hfPaper": "https://huggingface.co/papers/2608.11216", "paper": "https://arxiv.org/abs/2608.11216", "pdf": "https://arxiv.org/pdf/2608.11216", "project": "https://electronicarts.github.io/AutoWorldModelBench/", "report": "https://arxiv.org/abs/2608.11216"}, "metrics": [], "motivation": "World modeling is an unsettled field: architectures, training objectives, and state representations interact in complex ways, and no single recipe dominates across environments.", "name": "AutoWorldModel-Bench", "oneLine": "AutoWorldModel-Bench evaluates coding agents on autonomously improving a base world model under fixed compute budget across eight game environments. Uses structured-state representation, with held-out test split and closed-loop iteration.", "paperTitle": "AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.11216", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Electronic Arts", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://electronicarts.github.io/AutoWorldModelBench/"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.21483102985913266, "value": 4}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.47849787274907674, "value": 14}}, "confidence": "Medium", "coverage": 0.7, "rank": 204, "score": 34}, "method": "attention-ranking-v13", "rank": 204, "score": 34}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-20", "source": {"authors": ["Marjan Moodi", "Xuankang Zhu", "Fernando De Mesentier Silva", "Harold Chaput", "Mohammad Reza Taesiri"], "categories": ["cs.AI"], "id": "2608.11216", "title": "AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research", "type": "arxiv", "url": "https://arxiv.org/abs/2608.11216"}, "sourceUpdatedAt": "2026-08-16T01:00:16Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Addresses the gap in agent benchmarks for open-ended research tasks, enabling comparison of agents on research-like workflows rather than engineering-to-spec tasks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
