{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.15107", "kind": "paper", "url": "https://arxiv.org/abs/2606.15107"}, {"id": "gh:sanhornc/irts-toolbench", "kind": "repo", "url": "https://github.com/SanhornC/IRTS-ToolBench"}], "authors": ["Sanhorn Chen", "Xiaoyang Chen", "Boyu Liu", "Roy Zhao"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.15107", "source_url": "https://arxiv.org/abs/2606.15107"}, "categories": ["General AI", "Language & Knowledge", "Reasoning"], "description": {"en": "IRTS-ToolBench is a benchmark of 1,700 questions across 10 task types and 13 domains for evaluating irregular univariate time-series question answering, with standardized inputs and a reproducible evaluation protocol."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.15107", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.15107", "title": "IRTS-ToolBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of Illinois Urbana-Champaign"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.15107", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.15107", "source_url": "https://openalex.org/W7164939494"}, "key": "claire-radar:2606.15107", "modality": null, "name": "IRTS-ToolBench", "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_irts-toolbench_94993a2c", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "469cbe0966e38da01e197274f6b4245cb0a517605cf549d32a85ea9a2366ac1f", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.15107"}, "publisher": null, "released": "2026-06-13", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.15107", "source_url": "https://arxiv.org/abs/2606.15107"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-15107", "source": "claire_radar", "source_benchmark_id": "2606.15107", "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/SanhornC/IRTS-ToolBench", "githubScope": "benchmark_repo", "githubStars": 1, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.15107", "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": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "fresh"}, "capabilities": ["Reasoning"], "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": "7f0189ab569b5b0bf93070aa3fde64a64e744f7b6446d8edd1949923a10a6465", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "IRTS-ToolBench is a benchmark of 1,700 questions across 10 task types and 13 domains for evaluating irregular univariate time-series question answering, with standardized inputs and a reproducible evaluation protocol.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To bridge this gap, we introduce IRTS-ToolBench, a benchmark of 1,700 questions spanning 10 task types across 13 domains."}, "familyId": "bmf_cce157d7294f", "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/SanhornC/IRTS-ToolBench", "url": "https://github.com/SanhornC/IRTS-ToolBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/SanhornC/IRTS-ToolBench", "url": "https://huggingface.co/papers/2606.15107"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/SanhornC/IRTS-ToolBench", "url": "https://arxiv.org/abs/2606.15107"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/SanhornC/IRTS-ToolBench.", "url": "https://arxiv.org/abs/2606.15107"}], "scope": "benchmark_repo", "url": "https://github.com/SanhornC/IRTS-ToolBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_irts-toolbench_94993a2c", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/SanhornC/IRTS-ToolBench", "data": null, "hfPaper": "https://huggingface.co/papers/2606.15107", "paper": "https://arxiv.org/abs/2606.15107", "pdf": "https://arxiv.org/pdf/2606.15107", "project": null, "report": "https://arxiv.org/abs/2606.15107"}, "metrics": [], "motivation": "Time series data in real-world deployments is overwhelmingly irregular.", "name": "IRTS-ToolBench", "oneLine": "IRTS-ToolBench is a benchmark of 1,700 questions across 10 task types and 13 domains for evaluating irregular univariate time-series question answering, with standardized inputs and a reproducible evaluation protocol.", "paperTitle": "Towards Verifiable Agentic Data Science: Solving Irregular TSQA Via Tool-Grounded Reasoning", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.15107", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-13", "source": {"authors": ["Sanhorn Chen", "Xiaoyang Chen", "Boyu Liu", "Roy Zhao"], "categories": ["cs.AI"], "id": "2606.15107", "title": "Towards Verifiable Agentic Data Science: Solving Irregular TSQA Via Tool-Grounded Reasoning", "type": "arxiv", "url": "https://arxiv.org/abs/2606.15107"}, "sourceUpdatedAt": "2026-06-13T04:41:04Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Existing TSQA benchmarks assume regular sampling, leaving a gap for real-world irregular data. This benchmark provides standardized evaluation for LLMs and AI agents on irregular time series, with golden tool sets for tool-selection analysis."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
