{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.25819", "kind": "paper", "url": "https://arxiv.org/abs/2606.25819"}, {"id": "gh:foreverskyou/toolbench-x", "kind": "repo", "url": "https://github.com/Foreverskyou/ToolBench-X"}], "authors": ["Yang Tian", "Zhengpeng Shi", "Yu Zhou", "Bo Zhao"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.25819", "source_url": "https://arxiv.org/abs/2606.25819"}, "categories": ["Agents & Tool Use", "General AI", "cs.CL"], "description": {"en": "ToolBench-X evaluates tool-using agents on multi-step tasks with executable tools and automatic scoring, in environments that include recoverable reliability hazards such as specification drift, invocation errors, execution failures, output drift, and cross-source conflicts."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.25819", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.25819", "title": "ToolBench-X"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Shanghai Jiao Tong University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.25819", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.25819", "source_url": "https://openalex.org/W7165870665"}, "key": "claire-radar:2606.25819", "modality": null, "name": "ToolBench-X", "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_toolbench-x_cb76a568", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "6b325fc5644764426cfc59a1123fbed1683b1ac28e1219b5faf519146b643d3c", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.25819"}, "publisher": null, "released": "2026-06-24", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.25819", "source_url": "https://arxiv.org/abs/2606.25819"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-25819", "source": "claire_radar", "source_benchmark_id": "2606.25819", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Agents & Tool Use", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/Foreverskyou/ToolBench-X", "githubScope": "benchmark_repo", "githubStars": 3, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2606.25819", "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", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "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": "47490e3bfa5efc4a4f029089bdfb5b7c29fa6ed5d1611883b19f16895ce9da12", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ToolBench-X evaluates tool-using agents on multi-step tasks with executable tools and automatic scoring, in environments that include recoverable reliability hazards such as specification drift, invocation errors, execution failures, output drift, and cross-source conflicts.", "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": "We introduce ToolBench-X, a benchmark for evaluating agents under recoverable reliability hazards."}, "familyId": "bmf_da7a379f3f02", "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/Foreverskyou/ToolBench-X", "url": "https://github.com/Foreverskyou/ToolBench-X"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/Foreverskyou/ToolBench-X", "url": "https://huggingface.co/papers/2606.25819"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/Foreverskyou/ToolBench-X.", "url": "https://arxiv.org/abs/2606.25819"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/Foreverskyou/ToolBench-X", "url": "https://arxiv.org/abs/2606.25819"}], "scope": "benchmark_repo", "url": "https://github.com/Foreverskyou/ToolBench-X", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_toolbench-x_cb76a568", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/Foreverskyou/ToolBench-X", "data": null, "hfPaper": "https://huggingface.co/papers/2606.25819", "paper": "https://arxiv.org/abs/2606.25819", "pdf": "https://arxiv.org/pdf/2606.25819", "project": null, "report": "https://arxiv.org/abs/2606.25819"}, "metrics": [], "motivation": "Large language models are increasingly deployed as agents that solve tasks by interacting with external tool environments.", "name": "ToolBench-X", "oneLine": "ToolBench-X evaluates tool-using agents on multi-step tasks with executable tools and automatic scoring, in environments that include recoverable reliability hazards such as specification drift, invocation errors, execution failures, output drift, and cross-source conflicts.", "paperTitle": "Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.25819", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "ToolBench-X Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Foreverskyou/ToolBench-X"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-24", "source": {"authors": ["Yang Tian", "Zhengpeng Shi", "Yu Zhou", "Bo Zhao"], "categories": ["cs.CL", "cs.SE"], "id": "2606.25819", "title": "Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability", "type": "arxiv", "url": "https://arxiv.org/abs/2606.25819"}, "sourceUpdatedAt": "2026-06-27T07:38:41Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "Existing tool-use benchmarks assume stable tool environments, leaving a gap in evaluating agent performance under realistic unreliability. ToolBench-X provides a way to measure and compare agents' ability to diagnose and recover from tool hazards, which is crucial for practical deployment."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
