{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.08964", "kind": "paper", "url": "https://arxiv.org/abs/2607.08964"}, {"id": "gh:zli12321/lhtb", "kind": "repo", "url": "https://github.com/zli12321/LHTB"}, {"id": "hf:intelligencelab/long-horizon-terminal-bench", "kind": "dataset", "url": "https://huggingface.co/datasets/IntelligenceLab/Long-Horizon-Terminal-Bench"}], "authors": ["Zongxia Li", "Zhongzhi Li", "Yucheng Shi", "Ruhan Wang", "Junyao Yang", "Zhichao Liu", "Xiyang Wu", "Anhao Li", "Chengsong Huang", "Ergan Shang", "Yue Yu", "Ninghao Liu", "Lichao Sun", "Haotao Mi", "Leowei Liang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.08964", "source_url": "https://arxiv.org/abs/2607.08964"}, "categories": ["Agents & Tool Use", "Code", "General AI", "Multimodal", "Software & AI Compute"], "description": {"en": "Evaluates long-horizon terminal tasks in a containerized environment with hidden verifiers and dense reward grading across 46 tasks and nine categories."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.08964", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.08964", "title": "Long-Horizon-Terminal-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Tencent (China)", "Indiana University", "National University of Singapore", "University of Maryland, College Park", "University of Minnesota", "Hong Kong Polytechnic University", "Lehigh University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.08964", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.08964", "source_url": "https://openalex.org/W7168172066"}, "key": "claire-radar:2607.08964", "modality": null, "name": "Long-Horizon-Terminal-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_long-horizon-terminal-bench_593d14a6", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "1ec6c6c81f51a1901cfc94308677c5c78349afd5cb260ab2ce2598e5ea8169bb", "relation": "introduces", "review_state": "source-reviewed", "reviewed_at": "2026-08-30", "source_url": "https://arxiv.org/abs/2607.08964"}, "publisher": null, "released": "2026-07-09", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.08964", "source_url": "https://arxiv.org/abs/2607.08964"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-08964", "source": "claire_radar", "source_benchmark_id": "2607.08964", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI", "Software & AI Compute"], "area": "Agents & Tool Use", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/zli12321/LHTB", "githubScope": "benchmark_repo", "githubStars": 712, "hfDailySubmittedAt": "2026-07-13T00:00:00.000Z", "hfDataset": "IntelligenceLab/Long-Horizon-Terminal-Bench", "hfDatasetDownloads": 12351, "hfDatasetLikes": 137, "hfPaperUpvotes": 76, "hfPaperUrl": "https://huggingface.co/papers/2607.08964", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetLikes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "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": "77d3230de58f2233270e143a2d5ffcbf256c0414dba28bafbacab8bdb5e5148b", "model": "deepseek-v4-flash"}, "curation": {"reviewedAt": "2026-08-30", "sources": ["https://arxiv.org/abs/2607.08964", "https://github.com/zli12321/LHTB", "https://huggingface.co/datasets/IntelligenceLab/Long-Horizon-Terminal-Bench"], "state": "source-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates long-horizon terminal tasks in a containerized environment with hidden verifiers and dense reward grading across 46 tasks and nine categories.", "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 Long-Horizon-Terminal-Bench, a terminal benchmark of 46 long-horizon tasks spanning nine categories, including experiment reproduction, software engineering, multimodal analysis, interactive games, and scientific computing."}, "familyId": "bmf_0d2d7122f15f", "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/zli12321/LHTB", "url": "https://github.com/zli12321/LHTB"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/zli12321/LHTB", "url": "https://huggingface.co/papers/2607.08964"}], "scope": "project_repo", "url": "https://github.com/zli12321/LHTB", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_long-horizon-terminal-bench_593d14a6", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/zli12321/LHTB", "data": "https://huggingface.co/datasets/IntelligenceLab/Long-Horizon-Terminal-Bench", "hfPaper": "https://huggingface.co/papers/2607.08964", "paper": "https://arxiv.org/abs/2607.08964", "pdf": "https://arxiv.org/pdf/2607.08964", "project": null, "report": "https://arxiv.org/abs/2607.08964"}, "metrics": [], "motivation": "AI agents have become capable of autonomously completing short, well-specified tasks.", "name": "Long-Horizon-Terminal-Bench", "oneLine": "Evaluates long-horizon terminal tasks in a containerized environment with hidden verifiers and dense reward grading across 46 tasks and nine categories.", "paperTitle": "Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.08964", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"90d": {"confidence": "High", "coverage": 1.0, "datasetDownloadRank": 2, "datasetRankPopulation": 85, "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": 0.8769076758317191, "value": 712}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": 0.9702328890186308, "value": 12351}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.7675270026925342, "value": 76}}, "confidence": "High", "coverage": 1.0, "rank": 1, "score": 89}, "method": "attention-ranking-v13", "rank": 1, "score": 89}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-09", "source": {"authors": ["Zongxia Li", "Zhongzhi Li", "Yucheng Shi", "Ruhan Wang", "Junyao Yang", "Zhichao Liu", "Xiyang Wu", "Anhao Li", "Yue Yu", "Ninghao Liu", "Lichao Sun", "Haotao Mi", "Leowei Liang"], "categories": ["cs.AI"], "id": "2607.08964", "title": "Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading", "type": "arxiv", "url": "https://arxiv.org/abs/2607.08964"}, "sourceUpdatedAt": "2026-07-13T06:00:47Z", "topics": ["Multimodal", "Code"], "venueAttempts": [], "whyItMatters": "Provides a more demanding evaluation for agentic long-horizon planning and partial credit, addressing gaps in existing terminal benchmarks that only measure final outcomes."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
