{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.00267", "kind": "paper", "url": "https://arxiv.org/abs/2608.00267"}, {"id": "gh:microsoft/loopsbench", "kind": "repo", "url": "https://github.com/microsoft/Loopsbench"}, {"kind": "website", "url": "https://loopsbench.ai/"}], "authors": ["Han Li", "Zhemin Fang", "Rili Feng", "Yingqi Zhao", "Jiaheng Liu", "Pengfei Gao", "He Ye", "Dayi Lin", "Qingwei Lin", "Saravan Rajmohan", "Dongmei Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.00267", "source_url": "https://arxiv.org/abs/2608.00267"}, "categories": ["Agents", "Code & Software", "General AI"], "description": {"en": "LOOPSBENCH evaluates coding agents on long-horizon tasks structured as dependency DAGs with flow-aware test release and regression obligations, spanning 8 languages and 9 domains."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.00267", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.00267", "title": "LoopsBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Nanjing University", "Shanghai Jiao Tong University", "University College London"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.00267", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.00267", "source_url": "https://openalex.org/W7172416829"}, "key": "claire-radar:2608.00267", "modality": null, "name": "LoopsBench", "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-reviewed", "display_eligible": "unknown", "origin_record_id": "bm_loopsbench_86719709", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "7c80be963ccae454fecde4c84bec80b93aab6bf29591662bae872a1f28bbd4a7", "relation": "introduces", "review_state": "source-reviewed", "reviewed_at": "2026-08-19", "source_url": "https://arxiv.org/abs/2608.00267"}, "publication_dates": [{"basis": "paper_first_version", "date": "2026-07-31", "source_url": "https://arxiv.org/abs/2608.00267"}], "publisher": null, "released": "2026-07-03", "released_reference": {"basis": "first_public", "source_key": "claire-radar:2608.00267", "source_url": "https://github.com/microsoft/Loopsbench"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-00267", "source": "claire_radar", "source_benchmark_id": "2608.00267", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Code & Software", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/microsoft/Loopsbench", "githubScope": "benchmark_repo", "githubStars": 33, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 3, "hfPaperUrl": "https://huggingface.co/papers/2608.00267", "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": "05fe1d8410d59b6afe5cdae1ed2a975ac29afc278a5a0b431692ebd9672933aa", "model": "deepseek-v4-flash"}, "curation": {"reviewedAt": "2026-08-19", "sources": ["https://github.com/microsoft/Loopsbench", "https://loopsbench.ai/", "https://arxiv.org/abs/2608.00267"], "state": "source-reviewed"}, "dataStatus": "primary-source-reviewed", "demo": false, "description": "LOOPSBENCH evaluates coding agents on long-horizon tasks structured as dependency DAGs with flow-aware test release and regression obligations, spanning 8 languages and 9 domains.", "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 LOOPSBENCH, a long-horizon benchmark for loop engineering in coding agent evaluation."}, "familyId": "bmf_0c0fee9fe39e", "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/microsoft/Loopsbench", "url": "https://github.com/microsoft/Loopsbench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/microsoft/Loopsbench", "url": "https://huggingface.co/papers/2608.00267"}], "scope": "benchmark_repo", "url": "https://github.com/microsoft/Loopsbench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_loopsbench_86719709", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/microsoft/Loopsbench", "data": null, "hfPaper": "https://huggingface.co/papers/2608.00267", "paper": "https://arxiv.org/abs/2608.00267", "pdf": "https://arxiv.org/pdf/2608.00267", "project": "https://loopsbench.ai/", "report": "https://arxiv.org/abs/2608.00267"}, "metrics": [], "motivation": "Coding agent infrastructure is shifting from harness engineering toward loop engineering as coding agents are deployed for sustained long-horizon software development.", "name": "LoopsBench", "oneLine": "LOOPSBENCH evaluates coding agents on long-horizon tasks structured as dependency DAGs with flow-aware test release and regression obligations, spanning 8 languages and 9 domains.", "paperTitle": "LoopsBench: From Harness Engineering to Loop Engineering in Coding Agent Evaluation", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.00267", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Microsoft", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/microsoft/Loopsbench"}], "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": 0.4707057397524185, "value": 33}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.24495074076910422, "value": 3}}, "confidence": "Medium", "coverage": 0.7, "rank": 109, "score": 45}, "method": "attention-ranking-v13", "rank": 109, "score": 45}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releaseDates": {"firstPublicAt": "2026-07-03", "paperV1At": "2026-07-31"}, "releasedAt": "2026-07-31", "source": {"authors": ["Han Li", "Zhemin Fang", "Rili Feng", "Yingqi Zhao", "Jiaheng Liu", "Pengfei Gao", "He Ye", "Dayi Lin", "Qingwei Lin", "Saravan Rajmohan", "and Dongmei Zhang"], "categories": ["cs.SE", "cs.CL"], "id": "2608.00267", "title": "LoopsBench: From Harness Engineering to Loop Engineering in Coding Agent Evaluation", "type": "arxiv", "url": "https://arxiv.org/abs/2608.00267"}, "sourceUpdatedAt": "2026-08-10T12:29:11Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Provides a benchmark for loop engineering in sustained software development, assessing planning, implementation, and recovery over long horizons."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
