{"record": {"aliases": ["JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines"], "artifacts": [{"id": "arxiv:2606.19830", "kind": "paper", "url": "https://arxiv.org/abs/2606.19830"}], "authors": ["Jianwen Sun", "Chuanhao Li", "Zizhen Li", "Yukang Feng", "Fanrui Zhang", "Yifei Huang", "Yu Dai", "Kaipeng Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.19830", "source_url": "https://arxiv.org/abs/2606.19830"}, "categories": ["Code & Software", "General AI", "cs.SE"], "description": {"en": "JamBench evaluates project-level game code framework generation and code completion on 300 manually verified Godot projects using deterministic compilation, structural-completeness, and behavioral-alignment measures."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.19830", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.19830", "title": "JamBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.19830", "modality": null, "name": "JamBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-09-25T04:40:22Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_jamer-project-level-code-framework-dataset_d52443f9", "origin_source": "arxiv", "recognition_confidence": "0.6", "record_sha256": "7565875820eec9822066accf6c3a531548afbfb2110998707b33ab7481fae24d", "relation": "introduces", "review_model": "codex-cli-default", "review_state": "ai-reviewed", "reviewed_at": "2026-09-24T18:35:29.952059Z", "source_url": "https://arxiv.org/abs/2606.19830"}, "publisher": null, "released": "2026-06-18", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.19830", "source_url": "https://arxiv.org/abs/2606.19830"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-19830", "source": "claire_radar", "source_benchmark_id": "2606.19830", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Code & Software", "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses project-scale coding on professional game engines and reports large performance declines as project size increases.", "score": 7}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-24T18:35:29.952059Z", "inputHash": "b0d78ea74d411d21548884b5b412e5495639b69d59434b15e3a2bed1d6f19e61", "model": "codex-cli-default"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "JamBench, the first project-level game code framework dataset and benchmark built on a professional game engine.", "canonicalNameSource": "abstract", "decisionReason": "Publish because the abstract formally names JamBench, defines its tasks and scoring pipeline, describes deterministic verification, and states that its data and code are publicly available.", "evaluationEvidence": {"reuse": "All data and code are publicly available.", "scoring": "evaluated through a pipeline combining compilation pass rates, Structural Completeness Score (SCS), and Behavioral Alignment Score (BAS).", "task": "JamBench defines theme-driven generation and code completion tasks"}, "model": "codex-cli-default", "reviewedAt": "2026-09-24T18:35:29.952059Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "JamBench evaluates project-level game code framework generation and code completion on 300 manually verified Godot projects using deterministic compilation, structural-completeness, and behavioral-alignment measures.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We present JamSet and JamBench, the first project-level game code framework dataset and benchmark built on a professional game engine."}, "familyId": "bmf_9b78cea152f4", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_jamer-project-level-code-framework-dataset_d52443f9", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2606.19830", "pdf": "https://arxiv.org/pdf/2606.19830", "project": null, "report": "https://arxiv.org/abs/2606.19830"}, "metrics": [], "motivation": "Current AI-driven game development has made substantial progress in asset generation, gameplay design, and web-based game coding, yet project-level code engineering on professional game engines remains largely unexplored due to the absence of large-scale datasets and deterministic evaluation methods.", "name": "JamBench", "oneLine": "JamBench evaluates project-level game code framework generation and code completion on 300 manually verified Godot projects using deterministic compilation, structural-completeness, and behavioral-alignment measures.", "paperTitle": "JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.19830", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "readiness": "Paper only", "recognitionConfidence": 0.6, "relation": "introduces", "releasedAt": "2026-06-18", "reviewLifecycle": {"state": "pending"}, "source": {"authors": ["Jianwen Sun", "Chuanhao Li", "Zizhen Li", "Yukang Feng", "Fanrui Zhang", "Yifei Huang", "Yu Dai", "Kaipeng Zhang"], "categories": ["cs.SE", "cs.CL"], "id": "2606.19830", "title": "JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines", "type": "arxiv", "url": "https://arxiv.org/abs/2606.19830"}, "sourceUpdatedAt": "2026-06-21T05:48:00Z", "topics": ["cs.SE"], "venueAttempts": [], "whyItMatters": "It provides a repeatable evaluation of whether models can produce structurally complete, compilable, and behaviorally aligned code at the scale of professional game-engine projects."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
