{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.15242", "kind": "paper", "url": "https://arxiv.org/abs/2608.15242"}], "authors": ["Yunfei Zhang", "Boyu Feng", "Changhua Pei", "Zexin Wang", "Zhihuang Peng", "Xinlong Liu", "Hengyue Jiang", "Difeng Ma", "Jiayi Zhang", "Yongzhou Yao", "Yanan Zhao", "Fei Sun", "Yintong Huo", "Zhaoyang Liu", "Jingjing Li", "Gaogang Xie", "Dan Pei"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.15242", "source_url": "https://arxiv.org/abs/2608.15242"}, "categories": ["Agents", "General AI", "Language & Knowledge"], "description": {"en": "LongRCA Bench is a benchmark for diagnosing responsible roles and root causes in long-horizon agent failures. It comprises 1,140 failed trajectories across five domains, with human labels for the responsible role and the earliest decisive root-cause step. Evaluation focuses on responsible-role accuracy and exact root-step accuracy."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.15242", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.15242", "title": "LongRCA Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Chinese Academy of Sciences", "Computer Network Information Center", "Chongqing University", "University of Chinese Academy of Sciences", "Institute of Computing Technology", "Singapore Management University", "Alibaba Group (Cayman Islands)", "Alibaba Group (China)", "Tsinghua University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.15242", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.15242", "source_url": "https://openalex.org/W7203726780"}, "key": "claire-radar:2608.15242", "modality": null, "name": "LongRCA 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_longrca-bench_4c5e770a", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "d6c78de01d3c269e5333007db19edd97a74903499f82cacfdc5fcb3bb10ddf81", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.15242"}, "publisher": null, "released": "2026-08-15", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.15242", "source_url": "https://arxiv.org/abs/2608.15242"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-15242", "source": "claire_radar", "source_benchmark_id": "2608.15242", "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": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 15, "hfPaperUrl": "https://huggingface.co/papers/2608.15242", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "stale"}, "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": "18a8bafb2f851f51d696b408dd44815dc1f03e4f22efa0eeb56cff0438e9cde5", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "LongRCA Bench is a benchmark for diagnosing responsible roles and root causes in long-horizon agent failures. It comprises 1,140 failed trajectories across five domains, with human labels for the responsible role and the earliest decisive root-cause step. Evaluation focuses on responsible-role accuracy and exact root-step accuracy.", "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"], "snippet": "We introduce LongRCA Bench, comprising 1,140 failed trajectories across five domains without injected errors."}, "familyId": "bmf_a00231fd543d", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_longrca-bench_4c5e770a", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.15242", "paper": "https://arxiv.org/abs/2608.15242", "pdf": "https://arxiv.org/pdf/2608.15242", "project": null, "report": "https://arxiv.org/abs/2608.15242"}, "metrics": [], "motivation": "When a long-horizon agent execution fails, outcome-level evaluation reveals the unsuccessful result but not where the decisive error entered the trajectory.", "name": "LongRCA Bench", "oneLine": "LongRCA Bench is a benchmark for diagnosing responsible roles and root causes in long-horizon agent failures. It comprises 1,140 failed trajectories across five domains, with human labels for the responsible role and the earliest decisive root-cause step. Evaluation focuses on responsible-role accuracy and exact root-step accuracy.", "paperTitle": "LongRCA Bench: Diagnosing Responsible Roles and Root Causes in Long-Horizon Agent Failures", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.15242", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "LongRCA Bench Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2608.15242"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.15, "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": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.48990148153820845, "value": 15}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 50}, "method": "attention-ranking-v13", "rank": null, "score": 50}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-15", "source": {"authors": ["Yunfei Zhang", "Boyu Feng", "Changhua Pei", "Zexin Wang", "Zhihuang Peng", "Xinlong Liu", "Hengyue Jiang", "Difeng Ma", "Jiayi Zhang", "Yongzhou Yao", "Yanan Zhao", "Fei Sun", "Yintong Huo", "Zhaoyang Liu", "Jingjing Li", "Gaogang Xie", "Dan Pei"], "categories": ["cs.AI", "cs.SE"], "id": "2608.15242", "title": "LongRCA Bench: Diagnosing Responsible Roles and Root Causes in Long-Horizon Agent Failures", "type": "arxiv", "url": "https://arxiv.org/abs/2608.15242"}, "sourceUpdatedAt": "2026-08-15T14:06:02Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Long-horizon agent failures are difficult to debug, and outcome-level metrics obscure where errors occur. LongRCA Bench provides a standardized testbed for failure attribution, enabling comparison of methods that localize root causes and assign responsibility, which is crucial for improving agent reliability."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
